Economic and Social Reports
Women who work in seniors’ homes: Work-related injuries and illnesses and retention rates before the COVID-19 pandemic

Abstract

While a large body of research has identified several correlates of employee turnover, little is currently known regarding the degree to which employees who work in community care facilities for the elderly (industry group 6233 in the North American Industry Classification System [NAICS]), hereafter referred to as seniors’ homes, leave the nursing and residential care facilities subsector (NAICS 623) or the health care and social assistance sector (NAICS 62) in a given year. This study fills this information gap using data from the 2016 Census of Population integrated with Statistics Canada’s 2017 and 2018 Longitudinal Worker File, which enables us to link individuals selected in 2016 to their labour market status in the next two years. Sample size limitations restricted the focus of the study to women, who represent the vast majority—roughly 85%—of employees who work in seniors’ homes.

The study shows that up to 14% of women who worked in seniors’ homes in 2016 left the health care and social assistance sector that year, while up to 21% left the nursing and residential care facilities subsector. The leaving rate of women in seniors’ homes in 2016 is higher than the overall leaving rate of 12% observed when considering all industries. Women who earned relatively low wages, were not unionized or had low job tenure were generally more likely than others to leave. All else equal, immigrant women were less likely to leave than Canadian-born women.

The study also shows that in the three largest provinces (Quebec, Ontario and British Columbia), Canadian-born women employed as nurse aides in seniors’ homes were, all else equal, more likely to experience injuries or illnesses in 2016 than Canadian-born women who worked in different occupations for the same employer. In these three provinces, immigrant women employed as nurse aides in seniors’ homes were less likely to experience injuries or illnesses in 2016 than Canadian-born women employed as nurse aides for the same employer. However, regardless of immigrant status, province of residence or health sector considered, women who experienced injuries or illnesses in 2016 were no more likely than other women to leave health care. The study focuses on the pre-pandemic baseline because the 2022 and 2023 tax data are not available yet. When they become available, a post-pandemic period could be analyzed and compared to the pre-pandemic baseline.

Keywords: worker turnover, employee retention, quit rates, injury, illness, health care

Authors

Hanqing Qiu and René Morissette are with the Social Analysis and Modelling, Statistics Canada.

Acknowledgments

The authors wish to thank André Bernard, Feng Hou, Huda Masoud and Valentina Antonipillai for their comments on the first draft of this study.

Introduction

In the third quarter of 2015, the job vacancy rate in nursing and residential care facilities (subsector 623 in the North American Industry Classification System [NAICS]) was 1.9%, slightly below the rate of 2.6% observed nationwide (Chart 1). Eight years later, in the third quarter of 2023, 5.9% of jobs in nursing and residential care facilities were unfilled, a proportion that was almost 2 percentage points higher than the national job vacancy rate (4.1%).

These developments have raised concerns about employee retention in nursing care facilities (henceforth, nursing homes) and community care facilities for the elderly (henceforth, seniors’ homes), two industries that were heavily impacted by the COVID-19 pandemic (Clarke 2021). Given population aging, such concerns are important and unlikely to be short-lived.

While a large body of research has identified several correlates of employee turnover (Lazear and Oyer 2007; Hom et al. 2017; Trees Bolt, Winterton and Cafferkey 2022), little is currently known regarding the degree to which Canadian employees who work in seniors’ homes tend to leave the nursing and residential care facilities subsector (NAICS 623) or the health care and social assistance sector (NAICS 62) in a given year. The first goal of this study is to fill this information gap. Using data from the 2016 Census of Population and Statistics Canada’s Longitudinal Worker File (LWF), the study assesses the degree to which employees who worked in seniors’ homes in 2016 left the aforementioned subsector and sector that year—before the onset of the COVID-19 pandemic—and identifies which groups of employees had a relatively high likelihood of leaving.^Note  Sample size limitations restricted the focus of the study to women, who represent the vast majority—roughly 85%—of employees who work in seniors’ homes.

Data table for Chart 1



Data table for Chart 1
Table summary
This table displays the results of Data table for Chart 1 All industries and Nursing and residential care facilities, calculated using percent units of measure (appearing as column headers).

	All industries	Nursing and residential care facilities
	percent
2015
Q1	2.7	1.9
Q2	2.9	1.9
Q3	2.6	1.8
Q4	2.3	1.8
2016
Q1	2.2	1.8
Q2	2.5	1.8
Q3	2.6	1.9
Q4	2.4	1.9
2017
Q1	2.5	2.0
Q2	2.9	2.3
Q3	2.9	2.2
Q4	2.9	2.5
2018
Q1	2.9	2.5
Q2	3.4	2.9
Q3	3.3	3.1
Q4	3.3	3.0
2019
Q1	3.1	3.2
Q2	3.5	3.6
Q3	3.3	3.6
Q4	3.0	3.4
2020
Q1	3.1	3.7
Q2	Note ..: not available for a specific reference period	Note ..: not available for a specific reference period
Q3	Note ..: not available for a specific reference period	Note ..: not available for a specific reference period
Q4	3.5	5.2
2021
Q1	3.6	4.9
Q2	4.6	5.1
Q3	5.4	5.4
Q4	5.3	5.9
2022
Q1	5.2	6.5
Q2	5.9	6.8
Q3	5.6	7.7
Q4	4.8	7.0
2023
Q1	4.4	7.0
Q2	4.6	6.0
Q3	4.1	5.9
.. not available for a specific reference period Note: Data were not collected for the second and third quarters of 2020. Source: Statistics Canada, Table 14-10-0326-01.

The tasks performed by women employed in seniors’ homes and—more broadly—in the health care sector vary substantially across occupations. For example, nurse aides may need to reposition and transfer individuals during resident care, a task that puts them at greater risk of musculoskeletal injuries than other workers (Alamgir et al. 2007). Because occupational injuries and illnesses can potentially affect employee retention, it is important to document the degree to which they vary across occupations among employees who work in seniors’ homes. This is the second goal of the paper. Using descriptive evidence and multivariate analyses, the study sketches a profile of the women who are most likely to experience injuries or illnesses in a given year in the seniors’ homes industry group. The study also assesses whether the occurrence of injuries or illnesses is associated with an increased likelihood of women subsequently leaving the nursing and residential care facilities subsector or the health care and social assistance sector.

The paper proceeds as follows. The next section provides a brief literature review, followed by a description of the datasets and methodologies used. The third section examines the prevalence of injuries and illnesses in 2016 among various groups of employees who worked in seniors’ homes. The fourth section quantifies their rates of departure from various segments of the health care sector in 2016. Concluding remarks follow.

Background

A large literature in economics, management and psychology has documented several correlates of employee turnover over the last few decades. For example, Trees Bolt, Winterton and Cafferkey (2022) review 1,375 labour turnover studies published in 142 academic journals up to July 2019. Hom et al. (2017) review key publications on employee turnover during the 100-year existence of the Journal of Applied Psychology. Lazear and Oyer (2007) show how personnel economics—the application of economic and mathematical approaches to the study of human resource management—can shed light on several issues, such as compensation, the organization of work, and the matching process between firms and workers, all of which are closely related to employee turnover.

One key lesson learned from this body of research is that many factors—both monetary and non-monetary—appear to matter. For example, high-involvement work practices, control over one’s work environment, training opportunities, meaningfulness of the job, excessive or non-excessive workload, pace of work, quality of interpersonal relations with colleagues, trust in management, clarity of objectives, opportunities for career advancement within the organization, and perceptions of fairness (or lack thereof) are some of the numerous non-monetary factors that may potentially affect employee turnover.

As part of this large body of research, several Canadian studies have examined employee turnover intentions (e.g., Zeytinoglu et al. 2007; Zeytinoglu et al. 2009; Singh and Loncar 2010; Osuji et al. 2014; Lavoie-Tremblay et al. 2015; Fallatah, Laschinger and Read 2017; Fernet et al. 2017; Perreira, Berta and Herbert 2018; Zaheer et al. 2019; Nowrouzi-Kia and Fox 2020) or actual employee turnover (e.g., Denton et al. 2006; Rondeau, Williams and Wagar 2008, 2009; O’Brien-Pallas et al. 2010; Rondeau and Wagar 2012; Chu, Wodchis and McGilton 2014; Rondeau and Wagar 2016) in the health care sector, with most studies focusing on registered nurses. However, to the best of the authors’ knowledge, no Canadian study has documented the degree to which employees who work in seniors’ homes tend to leave various segments of the health care sector in a given year. Because employees in seniors’ homes represented 28% of all employees in nursing and residential care facilities in 2023 (as indicated by Labour Force Survey data), the lack of empirical evidence on labour turnover for this segment of the Canadian workforce is an important information gap. The first goal of this study is to fill it.

Employees in seniors’ homes work in a variety of occupations, such as administrative assistants, nurse aides, nurses, cleaners, kitchen helpers and social workers.^Note  The tasks performed by these different groups of workers vary substantially and may expose them to different risks of injuries and illnesses. For example, nurse aides must often reposition or transfer seniors to bathe them or help them move within a facility. Performing this task may put them at greater risk of musculoskeletal injuries, compared with other workers in the same establishment. For example, Alamgir et al. (2007) show that nurse aides who worked in British Columbia in community health care from 2004 to 2005 had higher injury rates than registered nurses. Whether nurse aides who work in seniors’ homes have relatively high rates of injury or illness in recent years or in other provinces remains an open question. The second goal of this study is to shed light on it.^Note 

Data, samples and methods

To achieve the two goals of this study, the 2016 Census of Population was integrated with Statistics Canada’s LWF. The LWF is a longitudinal dataset that tracks Canadian workers from 1989 to 2021 and allows for analyses of the degree to which employees who worked in a given industry, in a given year, left that industry subsequently. The LWF includes, among other variables, information on workers’ annual earnings, union status and coverage by a registered pension plan or deferred profit sharing plan, and an identifier for the enterprise in which workers are employed.^Note  However, it does not include information on workers’ education level, occupation or work hours. Because the 2016 Census of Population includes these variables, this limitation can be overcome by integrating the two files in the analysis.

The focus of the study is on employees who work in seniors’ homes (NAICS 6233). This includes employees who work in assisted living facilities with on-site nursing care, homes for the aged, old folks’ homes, old soldiers’ homes, retirement homes and rest homes, with health care incidental.

To quantify the degree to which employees in the seniors’ homes industry group in 2016 left the nursing and residential care facilities subsector (NAICS 623) or the health care and social assistance sector (NAICS 62) that year, the study uses a narrow sample and a broad sample.

The narrow sample consists of employees aged 18 to 60 who

1. were employed, in the LWF, as paid workers in NAICS 6233 in their main job—the job with the highest annual wages and salaries—in 2016
2. were paid workers in May 2016
3. had wages and salaries and positive weeks worked in 2015 but no income from self-employment in 2015
4. were permanent residents
5. lived in one of the 10 provinces in 2016
6. had knowledge of English, French or both languages in 2016
7. were not Indigenous people.

The broad sample consists of employees aged 18 to 60 who

1. were employed, in the LWF, as paid workers in NAICS 623 in any job in 2016
2. were permanent residents
3. lived in one of the 10 provinces in 2016
4. had knowledge of English, French or both languages
5. were not Indigenous people.

The two samples differ in important ways. The broad sample includes individuals employed in seniors’ homes in any job in 2016 and, therefore, includes jobs of short duration and jobs that ended early in the year. By contrast, the narrow sample restricts attention to individuals who were employed in this industry group in their main job. In addition, conditions 2 and 3 of the narrow sample imply that this sample consists of individuals who were employed as paid workers not only in May 2016 but also in 2015. Because the narrow sample focuses on the main job held in 2016 by individuals who had been employed for at least two years (instead of for only one year), it likely includes employees who had a higher degree of attachment to the labour market than those included in the broad sample. If so, the percentage of employees who leave the seniors’ homes industry group in a given year will likely be lower in the narrow sample than in the broad sample. The next section will investigate whether this is the case.^Note 

In both samples, roughly 85% of employees are women. Because sample sizes for men are often too small to support reliable multivariate analyses, the study restricts its focus to women. Results are shown separately for immigrant women and Canadian-born women because immigrant women—who represent about 35% of all women employed in seniors’ homes—may have fewer outside options than Canadian-born women and, therefore, may be less likely to leave the seniors’ homes industry group.

The study considers two outcomes.

The first is whether an employee experienced an injury or illness in 2016. This event is measured by a binary indicator that equals 1 if an employee satisfies at least one of the four following conditions in 2016: 1) received a record of employment because of injury or illness, 2) received workers’ compensation benefits, 3) received Canada Pension Plan or Quebec Pension Plan disability benefits and 4) had a disability tax credit for themselves or transferred this credit to a spouse. The percentages of employees experiencing an injury or illness in 2016—as measured by the binary indicator defined above—cannot be compared across provinces because the criteria used by provincial workers’ compensation boards to accept claims vary across provinces.^Note Note  However, these percentages can be compared across occupations within a given province. For this reason, the analyses of the occurrence of injuries and illnesses conducted in this study will be province-specific. Because of sample size limitations, such analyses will focus on the four largest provinces: Quebec, Ontario, Alberta and British Columbia.

The second outcome measures whether an employee in the seniors’ homes industry group left in 2016 1) the nursing and residential care facilities subsector (NAICS 623) or 2) the health care and social assistance sector (NAICS 62). In this study, an employee is deemed to have left the nursing and residential care facilities subsector in 2016 if she is not observed in NAICS 623 in 2017 and 2018, because she either moved to other industries, became unemployed or left the labour force. The second indicator applies a similar condition to NAICS 62. The rationale for both indicators is that transitions of employees outside the nursing and residential care facilities subsector and the health care and social assistance sector represent a loss of health care workers, narrowly or broadly defined.^Note 

Both outcomes are analyzed using descriptive evidence and logit models. In some cases, linear probability models that control for employer fixed effects are also used.

The study disaggregates the data into six occupational groups: 1) nurse aides, 2) nurses (registered and licensed), 3) social workers, 4) cleaners and kitchen helpers, 5) home support workers, and 6) other occupations. Throughout the study, the term “other occupations” is used specifically to refer to the sixth category. The terms “other employees” and “other women” are used to refer to all occupational groups, except the occupational group of interest. For instance, when nurse aides are the occupational group of interest, the two terms refer to categories 2 to 6.

Rates of injury and illness in 2016

Table 1 shows the percentage of women who experienced injuries or illnesses in 2016, as measured by the four conditions outlined in the previous section. Results are shown by occupation and various socioeconomic characteristics.

Of all immigrant women employed in seniors’ homes in 2016, about 8% to 9% experienced injuries or illnesses that year. Rates of injuries and illnesses for Canadian-born women varied from 11% to 14% and were, therefore, slightly higher.

Rates of injuries and illnesses varied substantially across occupations and education levels. For example, Canadian-born women employed as nurse aides—regardless of the sample considered—were twice as likely (17% to 19%) as their counterparts in other occupations (7% to 9%) to experience injuries or illnesses. Canadian-born women with a high school diploma or less education were also twice as likely as their counterparts with a bachelor’s degree or higher education to experience injuries or illnesses. Differences across occupations and education levels were less pronounced among immigrant women.



Table 1 also highlights large provincial differences in the percentage of women who experienced injuries or illnesses in 2016. These provincial differences cannot be easily interpreted because provincial workers’ compensation boards use different criteria for accepting claims submitted to them. For this reason, it is important to assess whether the cross-occupational differences in rates of injury or illness shown in Table 1 remain within provinces. This is done in Table 2 for the four largest provinces: Quebec, Ontario, Alberta and British Columbia.



The results indicate that, in all provinces, women who worked as nurse aides displayed higher rates of injury or illness than women who worked in other occupations. This was especially the case among Canadian-born women. For instance, rates of injury or illness among Canadian-born women employed as nurse aides in Quebec, Ontario and British Columbia were, in the broad sample, twice as high as those reported by women employed in other occupations. Once again, differences were more muted among immigrant women.

The relatively high rates of injury or illness reported by Canadian-born women employed as nurse aides in Quebec, Ontario and British Columbia are not solely attributable to group differences in socioeconomic characteristics. As Table 3 shows, most of the differences in rates of injury or illness observed between nurse aides and women employed in other occupations in these three provinces remain in logit models that control for an extensive set of covariates.^Note  For example, 21.9% of women employed as nurse aides in British Columbia in the broad sample experienced injuries or illnesses in 2016, a rate that exceeds the rate of 8.5% observed for their counterparts employed in other occupations by 13.4 percentage points (Table 2). Most of this difference (10.3 percentage points) remains in logit models, as the last column of Table 3 shows. The same pattern holds in Alberta, but the cross-occupational differences in rates of injury or illness between nurse aides and other occupations are smaller and estimated imprecisely in the logit models. For immigrant women, logit models estimated on the broad sample yield statistically significant differences between nurse aides and other occupations in the likelihood of experiencing injuries or illnesses only in Quebec.^Note 



Another possibility is that if nurse aides are overrepresented among organizations in which workers display relatively high rates of injury or illness, the relatively high rates of injury or illness reported by Canadian-born women employed as nurse aides in Quebec, Ontario and British Columbia might simply reflect this overrepresentation, instead of cross-occupational differences in rates of injury or illness within organizations.

The first panel of Table 4 dismisses this view. Even after controlling for employer fixed effects (using linear probability models) and workers’ age and education, about 90% of the differences in rates of injury or illness observed in Quebec, Ontario and British Columbia between nurse aides and other employees remains.^Note  Hence, most of these differences are observed within organizations.



Table 2 also shows that, regardless of the sample considered, the rates of injury or illness among nurse aides in Quebec and British Columbia are substantially lower among immigrant women than among Canadian-born women. For example, in the broad sample, the rates for British Columbia equal 13.3% among immigrant women and 21.9% among Canadian-born women. Part of the difference might reflect the fact that higher education is often associated with lower rates of injury or illness^Note  and that immigrant women employed as nurse aides are more educated than their Canadian-born counterparts.^Note  Another possibility is that immigrant nurse aides might work in organizations where rates of injury or illness are lower than organizations where Canadian-born nurse aides work.

The second panel of Table 4 tests these hypotheses. Even after controlling for employer fixed effects, and workers’ education and age, at least 80% of the differences in likelihood of experiencing an injury or illness—observed between immigrant and Canadian-born nurse aides who worked in Quebec, Ontario or British Columbia—remains. Thus, most of these differences cannot be accounted for by employer fixed effects or differences in educational attainment.^Note 

In the very narrowly defined occupation of nurse aides, immigrant women of a given age and educational attainment have lower rates of injury or illness than their comparable Canadian-born counterparts who worked for the same employer, and this raises important questions. One issue is whether the actual rates of injury or illness of the two groups are similar and whether immigrant women tend to claim employment insurance (EI) sickness benefits or workers’ compensation benefits to a lesser extent than Canadian-born women after an injury or illness. If so, a second issue concerns the factors that would lead injured immigrant women to claim benefits at a lower rate than their Canadian-born counterparts. Some of these factors may include reluctance to claim benefits because of cultural reasons, perceived job insecurity, the need to avoid work interruptions to send remittances abroad, and lack of awareness or understanding of the parameters governing EI sickness benefits and workers’ compensation (Kosny et al. 2012).^Note  Alternatively, if actual rates of injury or illness are lower among immigrant women than among Canadian-born women, then a natural question is which factors lead immigrant women employed as nurse aides in the seniors’ homes industry group to have relatively lower rates of injury and illness than their Canadian-born counterparts?^Note 

In summary, in the three largest provinces (Quebec, Ontario and British Columbia), Canadian-born women employed as nurse aides in the seniors’ homes industry group in 2016 experienced injuries or illnesses in greater proportions, all else equal, than other Canadian-born women who worked in that industry group. In these provinces, immigrant nurse aides had, all else equal, lower rates of injury or illness than Canadian-born nurse aides. Both patterns are observed within organizations, i.e., after controlling for employer fixed effects.

Women in the seniors’ homes industry group who left the nursing and residential care facilities subsector or the health care and social assistance sector in 2016

Table 5 presents the percentage of women in the seniors’ homes industry group who left the nursing and residential care facilities subsector (NAICS 623) or health care and social assistance sector (NAICS 62) in 2016. As noted above, a woman is deemed to have left the nursing and residential care facilities subsector in 2016 if she is not observed in NAICS 623 in 2017 and 2018, either because she moved to other industries or became non-employed. The second indicator applies a similar condition to NAICS 62.^Note 



Table 5 shows that these percentages vary substantially by immigrant status, sample type and health care subsector or sector considered. In 2016, 21.2% of Canadian-born women in the broad sample left the nursing and residential care facilities subsector. A smaller proportion (13.5%) left the health care and social assistance sector.^Note  In the narrow sample, the corresponding estimates drop by at least half and equal 8.8% and 6.6%, respectively. Hence, rates of departure are substantially lower in the narrow sample than in the broad sample.

Similar patterns are observed among immigrant women. However, the rates of departure of immigrant women are, in all cases, substantially lower than those of Canadian-born women. For instance, in the broad sample, 12.7% of immigrant women left the nursing and residential care facilities subsector in 2016, while 6.6% left the health care and social assistance sector that year.^Note 

Rates of departure also varied by occupation and province. In all samples, women who worked as nurse aides had lower rates of departure than those who worked in other occupations, even though the former group had higher rates of injury or illness than the latter. The relatively low rates of departure of nurse aides are observed for both immigrant and Canadian-born women. For all samples and health care subsectors or sectors considered, rates of departure of Canadian-born women were highest in Saskatchewan and lowest in Nova Scotia, with differences of at least 13 percentage points observed between the two provinces.

Notably, both immigrant and Canadian-born women employed as home support workers in seniors’ homes exhibited the highest rates of mobility from the nursing and residential care facilities subsector into other health care and social assistance industries, such as hospitals or home care. In the broad sample, 12.2% of immigrant home support workers and 12.4% of Canadian-born home support workers left the nursing and residential care facilities subsector in 2016 but remained employed within the health care and social assistance sector.

Other patterns are worth noting. Women who were under 25 years of age, earned relatively low wages, had no registered pension plan or deferred profit sharing plan in their job, were not unionized, or had low job tenure had higher rates of departure than other women. For example, 18.2% of recently hired Canadian-born women in the broad sample left the health care and social assistance sector in 2016, a rate that is 11.6 percentage points higher than that observed for their high-tenure counterparts, who had been with the same employer for at least six years (6.6%).

Most of these patterns hold in logit models that control for the socioeconomic characteristics displayed in Table 5. For example, the 11.6 percentage point difference in rates of departure from the health care and social assistance sector observed between recently hired Canadian-born women in the broad sample and their high-tenure counterparts drops to some extent but remains substantial, at 8.0 percentage points, in the logit models shown in Table 6. The 6.6 percentage point difference in rates of departure from nursing and residential care facilities observed in the narrow sample between unionized immigrant women and their non-unionized counterparts drops to 5.5 percentage points in logit models.



Cross-provincial differences in rates of departure are also observed among Canadian-born women in the broad sample in logit models. All else equal, Canadian-born women employed in the seniors’ homes industry group in 2016 in Nova Scotia were, in the broad sample, 11.0 percentage points less likely to leave the nursing and residential care facilities subsector that year than their counterparts employed in Quebec. This represents most of the 13.6 percentage point difference observed between these two groups in the raw data shown in Table 5.

By contrast, cross-occupational differences in the likelihood of leaving various health care subsectors or sectors remain in the broad sample, but often vanish in the narrow sample. For example, in the broad sample, Canadian-born women employed as nurse aides were less likely to leave various health care subsectors or sectors than Canadian-born women in other occupations. However, this is no longer the case in the narrow sample. 

Immigrant women versus Canadian-born women

Table 5 shows that, in general, immigrant women employed in the seniors’ homes industry group had lower rates of departure than Canadian-born women. Table 7 explores the sources of this difference. It uses linear probability models applied to samples of women who worked for employers that had, in the LWF–census data, at least one immigrant woman and at least one Canadian-born woman.



The first and second lines of Table 7 show that about half of the national differences in rates of departure between immigrant women and Canadian-born women can be accounted for by differences in the socioeconomic characteristics shown in Table 5. The third line shows that adding employer fixed effects to these socioeconomic characteristics does not further reduce the differences in rates of departure. This implies that the lower adjusted rates of departure of immigrant women are observed within organizations, rather than being attributable to their potential overrepresentation among employers that typically exhibit lower-than-average rates of departure.

Table 8 uses Oaxaca decompositions to identify which socioeconomic characteristics contribute the most to the national differences in rates of departure between these two groups of women. The last two columns show that in the broad sample, group differences in occupation, province of residence, union coverage and job tenure collectively account for about half of the national differences in rates of departure shown in Table 5. Specifically, the higher levels of job tenure of immigrant women, their higher union coverage, their overrepresentation in Ontario, and their overrepresentation as nurse aides and nurses—two occupations that display relatively low rates of departure—account for half of these national differences.^Note Note 



Broadly similar patterns are observed in the narrow sample. In this sample, group differences in the four variables above and in employment status and earnings collectively account for about half of the national differences in rates of departure.^Note 

The factors that explain the remaining differences in rates of departure are currently unknown. One possibility is that immigrant women who work in the seniors’ homes industry group live in low income in greater proportions than their Canadian-born counterparts and, as a result, choose to stay in their jobs in greater proportions. Census data do not provide strong support for this view. Results not shown indicate that in the narrow sample, after-tax low-income rates of immigrant women in 2015 were, at 2.8%, identical to those of Canadian-born women. In the broad sample, after-tax low-income rates were slightly higher among immigrant women who worked in the seniors’ homes industry group (6.8%) than among their Canadian-born counterparts (5.9%).^Note  Thus, there is little evidence that differences in low-income rates underlie the remaining differences in rates of departure.

Injuries and illnesses and rates of departure

Table 9 examines whether women who experienced injuries or illnesses in 2016 were more likely than others to leave the nursing and residential care facilities subsector or the health care and social assistance sector that year. Results are shown for the four largest provinces. In general, rates of departure from this subsector and sector were no higher for women who experienced injuries or illnesses than they were for other women. These findings are confirmed in multivariate analyses that control for all socioeconomic characteristics shown in Table 6 (except province of residence). Regardless of immigrant status, province of residence, or health subsector or sector considered, the likelihood of women with injuries or illnesses leaving is, all else equal, not statistically different from that of other women. This finding contrasts with what has been observed in some other sectors. Using 2016 LWF–census data, Morissette and Qiu (2023) show that early childhood educators and assistants (ECEAs) who experienced injuries or illnesses in 2016 were more likely than other ECEAs to leave the child care sector that year.



Concluding remarks

Using 2016 LWF–census data, this study documents the degree to which rates of injury or illness and rates of departure from health care subsectors or sectors differ for various groups of women employed in the seniors’ homes industry group in 2016. The main findings can be summarized as follows:

1. Up to 14% of women who worked in seniors’ homes in 2016 left the health care and social assistance sector that year. Up to 21% left the nursing and residential care facilities subsector.
2. In general, women who earned relatively low wages, were not unionized or had low job tenure were more likely than others to leave.
3. Rates of departure were higher among Canadian-born women than among immigrant women. About half of the difference in national rates of departure observed between these two groups can be accounted for by differences in occupation, job tenure, union coverage, province of residence, employment status and earnings.
4. In the three largest provinces (Quebec, Ontario and British Columbia), Canadian-born women employed as nurse aides were, all else equal, more likely to experience injuries or illnesses in 2016 than other Canadian-born employees. These differences are observed within organizations, i.e., even after controlling for employer fixed effects.
5. In these three provinces, immigrant women employed as nurse aides were, all else equal, less likely to report experiencing injuries or illnesses in 2016 than their Canadian-born counterparts. These differences also remain after accounting for employer fixed effects.
6. Regardless of immigrant status, province of residence, or health subsector or sector considered, women who experienced injuries or illnesses in 2016 were no more likely than other women to leave.

These findings raise a number of questions:

The first question is why immigrant nurse aides who work for a given employer experience, all else equal, injuries and illnesses to a lesser extent than Canadian-born nurse aides who work for the same employer. Is it because their actual rates of injury or illness are lower or because they sometimes refrain from claiming EI sickness benefits or workers’ compensation benefits for a variety of reasons?

A second question is why immigrant women leave the nursing and residential care facilities subsector or health care and social assistance sector in smaller proportions than Canadian-born women. Does this behaviour partly reflect intrinsic preferences for job stability, perceived economic insecurity or the need to sometimes send remittances abroad?

A third question is why immigrant and Canadian-born women who experienced injuries or illnesses in the seniors’ homes industry group in 2016 were no more likely than other women to leave various health care subsectors or sectors that year.

Lastly, the rates of departure shown in this study were measured in 2016, four years before the onset of the COVID-19 pandemic. One would expect these rates of departure to be higher during and after the COVID-19 pandemic. Whether this is the case remains to be examined when the 2022 and 2023 LWF become available by the end of 2025.

The investigation of these four questions is left for future research.

References

Alamgir, H., Y. Cvitkovich, S. Yu and A. Yassi. 2007. “Work-related injury among direct care occupations in British Columbia, Canada.” Occupational and Environmental Medicine, 64(11): 769-775.

Chowhan, J., M. Denton, C. Brookman, S. Davies, F.K. Sayin and I.U. Zeytinoglu. 2019. “Work intensification and health outcomes of health sector workers.” Personnel Review, 48(2), 342-359.

Chu, C.H., W.P. Wodchis and K.S. McGilton. 2014. “Turnover of regulated nurses in long-term care facilities.” Journal of Nursing Management, 22(5), 553-562.

Clarke, J. 2021. “Impacts of the COVID-19 pandemic in nursing and residential care facilities in Canada.” StatCan COVID-19: Data to Insights for a Better Canada, Statistics Canada Catalogue no. 45280001.

Denton, M., I.U. Zeytinoglu, S. Davies and D. Hunter. 2006. “The impact of implementing managed competition on home care workers’ turnover decisions.”  Healthcare Policy, 1(4), 106-123.

Drebit, S., S. Shajari, H. Alamgir, S. Yu and D. Keen. 2010. “Occupational and environmental risk factors for falls among workers in the healthcare sector.” Ergonomics, 53(4), 525-536.

Fallatah, F., H.K.S. Laschinger and E.A. Read. 2017. “The effects of authentic leadership, organizational identification, and occupational coping self-efficacy on new graduate nurses’ job turnover intentions in Canada.” Nursing Outlook, 65(2), 172-183.

Fernet, C., S.-G. Trépanier, M. Demers and S. Austin. 2017. “Motivational pathways of occupational and organizational turnover intention among newly registered nurses in Canada.” Nursing Outlook, 65(4), 444-454.

Hom, P.W., T.W. Lee, J.D. Shaw and J.P. Hausknecht. 2017. “One hundred years of employee turnover theory and research.” Journal of Applied Psychology, 102(3), 530-545.

Houle, R. and G. Schellenberg. 2008. “Remittances by recent immigrants.” Perspectives on Labour and Income, Statistics Canada Catalogue no. 75-001-X.

Kosny, A., M. Lifshen, E. MacEachen, P. Smith, G.J. Jafri, C. Neilson, D. Pugliese and J. Shields. 2012. “Delicate dances: immigrant workers’ experiences of injury reporting and claim filing.” Ethnicity & Health, 17(3), 267-290.

Lavoie-Tremblay, M., C. Fernet, G.L. Lavigne and S. Austin. 2015. “Transformational and abusive leadership practices: impacts on novice nurses, quality of care and intention to leave.” Journal of Advanced Nursing, 72(3), 582-592.

Lazear, E. and P. Oyer. 2007. “Personnel economics.” National Bureau of Economic Research Working Paper no. 13480.

Morissette, R. and H. Qiu. 2023. “Work absences due to injury or illness and employee retention in the child care services industry before the COVID-19 pandemic.” Analytical Studies Branch Research Paper Series no. 471, Statistics Canada Catalogue no. 11F0019M.

Nowrouzi-Kia, B. and M.T. Fox. 2020. “Factors associated with intent to leave in registered nurses working in acute care hospitals: a cross-sectional study in Ontario, Canada.” Workplace Health and Safety, 68(3), 121-128.

O’Brien-Pallas, L., G.T. Murphy, J. Shamian, X. Li and L.J. Hayes. 2010. “Impact and determinants of nurse turnover: a pan-Canadian study.” Journal of Nursing Management 18, 1073-1086.

Osuji, J., F.M. Uzoka, F. Aladi and M. El-Hussein. 2014. “Understanding the factors that determine registered nurses’ turnover intentions.” Research and Theory for Nursing Practice, 28(2), 140-161.

Perreira T.A., W. Berta and M. Herbert. 2018. “The employee retention triad in health care: exploring relationships amongst organisational justice, affective commitment and turnover intention.” Journal of Clinical Nursing, 27(7-8), 1451-1461.

Premji, S., K. Lippel and K. Messing. 2008. “‘On travaille à la seconde!’ Rémunération à la pièce et santé et sécurité du travail dans une perspective qui tient compte de l’ethnicité et du genre.” Perspectives interdisciplinaires sur le travail et la santé, (10-1).

Rondeau, K.V., E.S. Williams and T.H. Wagar. 2008. “Turnover and vacancy rates for registered nurses: do local labor market factors matter?” Health Care Management Review, 33(1), 69-78.

Rondeau, K.V., E.S. Williams and T.H. Wagar. 2009. “Developing human capital: what is the impact on nurse turnover?” Journal of Nursing Management, 17(6), 739-748.

Rondeau, K.V. and T.H. Wagar. 2012. “Employee high-involvement work practices and voluntary turnover: does human capital accumulation or an employee empowerment culture mediate the process? Examining the evidence in Canadian healthcare organizations.”  European Conference on Intangibles and Intellectual Capital; Kidmore End: Academic Conferences International Limited, 386-394.

Rondeau, K.V. and T.H. Wagar. 2016. “Human resource management practices and nursing turnover.” Journal of Nursing Education and Practice, 6(10), 101-109.

Scherzer, T., R. Rugulies and N. Krause. 2005. “Work-related pain and injury and barriers to workers’ compensation among Las Vegas hotel room cleaners.” American Journal of Public Health, 95(3), 483-488.

Singh, P. and N. Loncar. 2010. “Pay satisfaction, job satisfaction and turnover intent.” Industrial Relations, 65(3), 470-490.

Smith, P.M., C. Chen and C. Mustard. 2009. “Differential risk of employment in more physically demanding jobs among a recent cohort of immigrants to Canada.” Injury Prevention, 15(4), 252-258.

Trees Bolt, E.E., J. Winterton and K. Cafferkey. 2022. “A century of labour turnover research: a systematic literature review.” International Journal of Management Reviews, 24(4), 555-576.

Yassi, A., M. Gilbert and Y. Cvitkovich. 2005. “Trends in injuries, illnesses, and policies in Canadian healthcare workplaces.”  Canadian Journal of Public Health, 96(September), 333-339.

Zaheer, S., L. Ginsburg, H.J. Wong, K. Thomson, L. Bain and Z. Wulffhart. 2019. “Turnover intention of hospital staff in Ontario, Canada: exploring the role of frontline supervisors, teamwork, and mindful organizing.” Human Resources for Health, 17, 66.

Zeytinoglu, I.U., M. Denton and S. Davies. 2002. “Casual jobs, work schedules and self-reported musculoskeletal disorders among visiting home care workers.” Unpublished.

Zeytinoglu, I.U., M. Denton, S. Davies, A. Baumann, J. Blythe and L. Boos. 2007. “Deteriorated external work environment, heavy workload and nurses’ job satisfaction and turnover intention.” Canadian Public Policy, 33(s1), 31-48.

Zeytinoglu, I.U., M. Denton, S. Davies and J.M. Plenderleith. 2009. “Casualized employment and turnover intention: home care workers in Ontario, Canada.” Health Policy, 91(3), 258-268.