Paid sick leave benefits among essential frontline workers serving people experiencing homelessness in Canada during the COVID-19 pandemic.

OBJECTIVES: This study examined the prevalence and factors associated with paid sick leave benefits among direct service providers who work with people experiencing homelessness. STUDY
DESIGN: Cross-sectional study using an online survey disseminated during the second wave of the COVID-19 pandemic in Canada.
METHODS: Survey data from 572 direct service providers working in the homeless, supportive housing, and harm reduction service sectors were analyzed for this study. Univariate and multivariate logistic regression models were used to examine predictors of paid sick leave benefits.
RESULTS: One hundred one (17.7%) participants did not have any paid sick leave benefits. In the univariate models, paid sick leave was associated with older age, greater family income, full-time work, specific employment settings (supportive housing and not emergency shelters or harm reduction programs), having a regular medical doctor, and fewer occupational impacts of the COVID-19 pandemic. Older age, full-time work, and non-receipt of emergency financial benefits remained statistically significant predictors in the multivariate model.
CONCLUSIONS: Although the majority of service providers working with people experiencing homelessness have some amount of paid sick leave benefits, there is a precariously employed subset of individuals who are younger and working part-time in the sector. Temporary expansion of paid sick leave and removal of waiting periods for new employees to qualify for benefits are recommended.

Introduction

Paid sick leave has been identified as a key public health strategy for reducing transmission of COVID-19. However, in Canada, over 40% of adults do not receive paid sick leave benefits from their employers. In the context of the COVID-19 pandemic, this can force people to make difficult decisions between protecting public health or financially supporting themselves and their families. Essential frontline workers without paid sick leave face the heightened precarity of being designated as needed by society (or, the laws enacted during the pandemic) and, thus, are expected to do their jobs without a protection that could keep them healthy and safe. As some essential frontline workers serve vulnerable populations, the consequences of the lack of paid sick leave benefits can be deleterious not only to workers but also those receiving care and support.

People experiencing homelessness are one population that is highly vulnerable to COVID-19. A recent study found that people with recent histories of homelessness were over 3 times more likely to have a positive test result for COVID-19, 20 times more likely to be admitted to hospital for the virus, 10 times more likely to require intensive care for the virus, and 5 times more likely to die within the first 21 days of a positive test than people with stable housing. One contributing factor to their heightened risk is the congregate nature of emergency shelters, which can be crowded and make other public health measures, such as physical distancing, more challenging. Research has shown that these communal living settings are susceptible to outbreaks, which can cause rapid transmission of the virus to large proportions of service users and providers. , Yet, despite paid sick leave being an important public health intervention during the pandemic, there is a dearth of evidence on these benefits within the workforce that serves the homeless population. Understanding the prevalence and factors associated with paid sick leave benefits among this workforce could help to reduce COVID-19 transmission and spreading within the medically vulnerable homeless population. Accordingly, this study examined the prevalence and factors associated with paid sick leave among direct service providers who work with people experiencing homelessness in Canada.

Methods

As part of an ongoing study of the mental health and well-being of direct service providers working with people experiencing homelessness in Canada, an online survey was developed and disseminated widely across the country via emails to sector-specific networks and organizations, newsletters, listservs, social media, and word of mouth. The survey collected data on background and occupational information, including paid sick leave benefits; health and well-being; and perceived impacts of the COVID-19 pandemic. Service providers were eligible to participate in the study if they [a] were 18 years of age or older; [b] worked in Canada; [c] provided direct services to people experiencing homelessness; and [d] worked in homeless (including specialized health services for people experiencing homelessness), supportive housing, or harm reduction services. The study was approved by the institutional review board of the lead author's affiliation.

A total of 579 participants completed the online survey between November 12 and December 22, 2020—a six-week period during the second wave of the COVID-19 pandemic in Canada that had a daily average of 5949.07 cases and 91.46 deaths. Seven (1.2%) participants were removed from analysis due to missing data on sick leave benefits. Univariate and multivariate logistic regression models were used to examine the predictors of paid sick leave benefits among direct service providers working with people experiencing homelessness. All variables that were statistically significant in the univariate models were entered into the multivariate model.

Results

The sample was similar to the homeless service sector workforce in gender (female: 79.9%; male: 16.6%; transgender/nonbinary: 2.8%), age (M = 39.05 years, SD = 12.60), ethnicity (white: 80.4; non-white: 18.4%). Forty-four participants (7.7%) identified as being Indigenous. Lived experience of behavioral health problems (57.0%) and homelessness (15.9%) were also common among direct service providers. All provinces and territories except the Yukon were represented in the sample; 9.1% were from the Atlantic region, 57.9% were from Central Canada, 15.1% were from the Prairie Provinces, 16.6% were from the West Coast, and 1.2% were from the Northern Territories. Most participants (70.1%) worked in urban settings, whereas 29.4% provided services to at least one rural, remote, or small community.

Of the 572 participants, 101 (17.7%) had no paid sick leave benefits, 121 (21.2%) had one week or less, and 350 (61.2%) had more than one week. The univariate and multivariate models predicting paid sick leave benefits (any amount) are shown in Table 1 . In the univariate models, paid sick leave benefits were positively associated with older age, greater family income, full-time work, employment in supportive housing, and having a regular medical doctor. By contrast, service providers without paid sick leave benefits were more likely to be working in emergency shelters and harm reduction programs, have had their work hours decreased during the COVID-19 pandemic, and have accessed pandemic-specific emergency financial benefits (i.e. Canada Emergency Response Benefit). In the multivariate model, three factors remained significant predictors of paid sick leave benefits: older age, full-time employment, and non-receipt of emergency financial benefits.

Table 1

Predictors of paid sick leave benefits among 572 direct service providers working with people experiencing homelessness.

Predictor	Univariate models			Multivariate model
	UOR	95% CI	p	AOR	95% CI	p
Gender
Male (reference)	1.00
Female/transgender/non-binary	0.69	0.37–1.30	0.25
Age	1.04	1.02–1.06	<.001	1.04	1.02–1.07	.001
Ethnicity
White (reference)	1.00
Non-white/mixed race	1.06	0.61–1.86	0.83
Education
High school or less (reference)	1.00			1.00
College diploma	2.34	1.07–5.13	.03	2.56	0.90–7.25	0.08
Bachelor's degree	1.74	0.83–3.62	0.14	2.34	0.86–6.37	0.10
Graduate degree	2.24	0.87–5.79	0.10	2.42	0.69–8.44	0.17
Family income
$39,999 or less (reference)	1.00			1.00
$49,000–59,999	5.14	2.66–9.95	<.001	2.11	0.91–4.92	0.08
$60,000–79,999	2.55	1.23–5.16	<.01	0.94	0.39–2.28	0.90
$80,000–99,999	2.69	1.31–5.52	<.01	1.31	0.52–3.28	0.57
$100,000 or more	3.24	1.73–6.04	<.001	1.35	0.58–3.14	0.49
Work amount
Part-time (reference)	1.00			1.00
Full-time	13.41	8.15–22.06	<.001	15.54	8.52–28.34	<.001
Lived experience of homelessness	1.10	0.61–2.01	0.75
Lived experience of behavioral health problems	1.19	0.77–1.83	0.43
Work setting
Emergency shelter	0.57	0.36–0.89	.02	0.64	0.33–1.26	0.20
Supportive housing	2.03	1.18–3.51	.01	1.38	0.64–2.99	0.41
Community-based health	1.69	0.89–3.22	0.11
Harm reduction	0.40	0.21–0.75	<.01	0.73	0.28–1.88	0.51
Have a regular medical doctor	1.75	1.02–3.00	.04	1.32	0.65–2.68	0.44
Unmet behavioral health needa	0.94	0.53–1.65	0.83
COVID-19 impacts
Worsened mental health	1.07	0.63–1.82	0.80
Increased stress	1.78	0.88–3.58	0.11
Increased alcohol use	0.83	0.51–1.33	0.43
Increased cannabis use	0.83	0.50–1.37	0.46
Decreased work hours	0.29	0.18–0.48	<.001	0.57	0.29–1.13	0.11
Accessed financial benefitsb	0.26	0.16–0.44	<.001	0.35	0.17–0.72	<.01

UOR = unadjusted odds ratio; AOR = adjusted odds ratio; CI = confidence interval.

Past 12 months.

A national, emergency financial benefit for workers impacted by the pandemic.

Discussion

The results highlight that most service providers working with people experiencing homelessness have some amount of paid sick leave benefits. However, there is a precariously employed subset of service providers who are younger and working part-time in the sector. This group was also more likely to have accessed governmental emergency financial benefits, suggesting that the pandemic has had a greater financial toll on those without paid sick leave. As emergency shelters and harm reduction programs were negatively associated with paid sick leave benefits in the univariate models, but not the multivariate one, this is likely due to these settings having a higher proportion of part-time workers (emergency shelters: 22.6% part-time workers; harm reduction programs: 35.3% part-time workers; full sample: 19.0% part-time workers). Given that these services provide support to groups that are medically vulnerable, policies are needed to temporarily expand paid sick leave benefits to service providers working in these settings for the duration of the pandemic. Furthermore, as emergency shelters are vulnerable to COVID-19 outbreaks, preventive measures, such as paid sick leave, are critical to protecting service users and providers. Similarly, removal of waiting periods to qualify for paid sick leave benefits for new employees is strongly recommended. Those without paid sick leave in this study were also more likely to draw on these financial benefits. Given this, providing paid sick leave to the precariously employed subset of service providers without benefits could potentially offset the costs of governmental emergency financial benefits.

There are several notable study limitations. First, few participants were from Quebec, so the findings may be less generalizable to service providers working in that province. Second, the sample was more educated than the homeless service sector workforce. Third, as the online survey was disseminated via homeless service networks and organizations, part-time employees without paid sick leave benefits, especially those who had their work hours reduced, may have been less able to participate in the study. The latter two limitations may have contributed to a higher rate of paid sick leave benefits in the sample. Nevertheless, the findings highlight a critical gap in paid sick leave benefits within an essential frontline workforce that needs to be addressed to reduce risks of COVID-19 transmission and spreading within the homeless population.

Author statements

Ethical approval

Centre for Addiction and Mental Health (#085/2020).

Funding

(MFE-171228) and the Discovery Fund.

Competing interests

None declared.