Abstract
Women are the fastest growing population in federal prison in Canada. Women's incarceration has significant implications for families, as approximately two-thirds have children who face intergenerational trauma, risk of criminalization, and health concerns. The Correctional Services Canada Mother Child Program allows children up to age six to live with their incarcerated mothers. Publicly available information about outcomes associated with the program is scarce, including the number of participants. Using data from 2000–2020 acquired through an Access to Information and Privacy request, this article presents descriptive statistics about the program. Findings indicate the program is underused, and associated outcomes are under-researched.
Keywords
Introduction
Women are the fastest growing population in prisons in Canada, with the number of federally sentenced women increasing 16.3% from 2010 to 2019 (Public Safety Canada, 2019). Women are a small subset of the total incarcerated population, representing 6% of federally sentenced people (Correctional Services Canada, (CSC), 2019). Federal prisoners have received a sentence of two years or more, whereas those in provincial institutions are either in pre-trial custody or have a sentence of less than 2 years. Incarcerated people have complex social and health histories including childhood abuse, sexual trauma, PTSD, substance use disorder, mental illness, chronic illness, and infectious disease (Kouyoumdjian et al., 2016). They also experience high rates of unemployment, low educational attainment, and low literacy (Mahony et al., 2017). Women's incarceration has significant implications for families, as approximately two-thirds are mothers (Mallicoat, 2014) with an average of four children (Kouyoumdjian et al., 2016). These children face intergenerational trauma (Mussell, 2020), and risk criminalization themselves (Withers & Folsom, 2008), as well as multiple health concerns and premature death (Centre for Disease Control and Prevention (CDC), 2019; Felitti et al., 1998).
In 1990, the Taskforce on the Future of Federally Sentenced Women (TFFSW) released a report called Creating Choices that called for the creation of environments appropriate for children to live with their mothers (TFFSW, 1990). Formally implemented across the federal system in 2001 and governed by Commissioner's Directive 768 (CD-768), the stated purpose of CSC's Mother Child Program (MCP) is “to foster positive relationships between women incarcerated in women offender institutions and units and their children and to provide a supportive environment that promotes stability and continuity for the mother-child relationship” (CSC, 2020, p. 1). Mothers who have not been convicted of an offense against a child, are classified as minimum or medium security risk, and have children up to the age of six may apply (CSC, 2020). Program supporters point to evidence that keeping mothers and children together prevents psychological, physiological and developmental harm to the child (The Collaborating Centre for Prison Health and Education, 2015). Abolitionist theorists would counter that prisons are high-risk environments, with elevated risks of disease transmission, injury and death, and the high cost of operating prison programs such as the MCP displaces funding that could be directed towards community-based solutions (Critical Resistance-INCITE, 2003; Davis, 2003; Kaba, 2021; Gilmore, 2007). Publicly available information about outcomes associated with the federal MCP is scarce, including the number of program participants. This article begins to illuminate MCP participation.
Commissioner's Directive (CD) 768 sets out the responsibilities, eligibility, requirements and processes for the MCP (CSC, 2020). Each of the six federal facilities for women has an MCP (Table 1). There is also a small 28-bed healing lodge for federally sentenced women that opened in 2011 and is operated by Native Counselling Services Alberta (Native Counselling Services Alberta, n.d.). In 2014, CSC added 114 minimum security beds to its facilities for women, and 15 new rooms specifically for the MCP (The Canadian Press, 2014).
Capacity and MCP Participants at Federal Prisons for Women.
Buffalo Sage opened in 2011.
There is one provincial facility with an MCP at the Alouette Correctional Centre for Women in Maple Ridge, BC. ACCW is a large facility with capacity for 315 women (Paynter, Bagg, et al., 2020). From 2004–2008, 12 women gave birth while incarcerated and kept their infants with them. In 2008, the province closed the program, which prompted a constitutional challenge by several of the women. In 2013, British Columbia Supreme Court Justice Carol Ross ruled that restricting mothers and babies from being together violated the Canadian Charter of Rights and Freedoms (Inglis v BC, 2013). Since the program reopened in 2014, there have been two participants (Minister for Public Safety and Solicitor General Corrections Branch, 2016). No other province has started a program.
In 2019, the OCI, the ombudsperson for federally incarcerated people, stated that 125 applications to the MCP had ever been submitted, and of those, 108 were successful (Canadian Friends Service Committee, 2019). Considering that, on average, 701 women are federally incarcerated on a given day (Public Safety, 2019), and approximately two-thirds of them are mothers, the number of applications is strikingly low. In this article we examine available data on MCP participation in Canada.
Literature Review
In an international policy review, Robertson (2012) found most countries around the world allow children to live with their mothers in prison. However, the infrastructure, program elements, and eligibility requirements of MCPs varied considerably around the world. Some prisons locate the MCP in a separate unit from the general population, as is the case for most of the MCPs in England. In New Zealand, mothers and children live together in special facilities called “self care units” (Robertson, 2012, p. 22). Describing the MCP in an Italian prison, Candelori and Del Rossi (2007) write: “The “nursery” has all the features of a prison ward: barred windows, steel doors, locks, gates with iron bars. Even the garden-courtyard is surrounded by a high fence.” (p. 61). In some of the federal prisons for women in Canada, MCP participants live apart from the general population in the Minimum-Security Unit, an apartment-building-style facility outside the gates (Paynter, 2021). Some jurisdictions include day care centers, such as the Ohio Reformatory for Women (Inskeep, 2008). The Creating Choices report in Canada called for daycare in the prisons (Task Force on Federally Sentenced Women, 1990), but in the long term this did not materialize.
Most jurisdictions place age restrictions on child eligibility. In Germany children can live full-time inside until they are school-age or about 5 years old (Robertson, 2012). In the US, the United Kingdom and many programs on continental Europe, the age restrictions are much earlier, around 18 months. Black et al. (2004) have criticized age limits as arbitrary and without grounding in evidence.
There are a few studies of how MCP participation supports mothers in maintaining long term custody of their children. In their US-based case-control study, Pimlott Kubiak et al. (2010) found that a greater proportion of children of the MCP intervention group had contact with child protection services (CPS) than in the control group. In a study of 100 infants in an MCP, Byrne et al. (2012) found the most common reasons for a child to be separated from their mother after participation was due to the child being removed by CPS or the child reaching the program age limit. Among the children in the French study by Blanchard et al. (2018), at discharge from the program, 40% of the children were removed from their mothers by CPS. As Haney (2013) explains, “what began as a promising alternative to punishment morphed into its own form of power and control” (p.107).
Canada
The federal MCP in Canada is decidedly under-researched. Paynter, Jefferies, et al. (2020) published a scoping review of the health outcomes associated with participation in MCPs around the world. The review included 27 studies; however, the authors did not find any relevant studies in Canada. A small number of social science and law articles did examine barriers to MCP participation in Canada. Brennan (2014) found declining participation in the MCP from 2001–2012 to be related to: 1) changes to MCP eligibility criteria enacted in 2008 that made participation more exclusive; 2) overcrowding in the prison facilities, a result of increasing incarceration of women; 3) limitations of the physical prison environment and 4) the increasingly punitive nature of corrections.
Miller (2017) considers how CSC policies and practices create barriers to MCP participation for Indigenous women. The Custody Rating Scale used in security classification includes factors such as “employment, marital status, family situation, associates, social interaction, substance abuse, community functioning, personal and emotional orientation and attitude” (Miller, 2017, p. 12). People with higher security classification are excluded from potentially helpful programs such as the MCP and experiences of colonial harm result in elevated classification among Indigenous women. Further, Miller asserts that Indigenous women are less likely to accept the involvement of CPS, a requirement for MCP participation, because Indigenous children are more likely to be removed from their families. It is estimated that 52.2% of children in care under the age of 14 are Indigenous (Government of Canada, 2020). Fayter and Payne (2017) recount the story of a federally incarcerated woman at Grand Valley Institution who wanted to participate in the MCP. Despite “being a model prisoner” (p.16), and having received approval from the CPS, she was consistently denied the necessary security reclassification required.
In general, the literature review documents a lack of evidence or theory governing the programs and a lack of comparative analysis between programs to determine best practices. The assumption that the MCP in Canada promotes mother-child connections should be scrutinized in the long-term.
Method
As part of a larger program of research to examine maternal and reproductive health services for individuals incarcerated in prisons for women in Canada, this article presents the quantitative findings about MCP participation. This program of research uses a theoretical framework of feminist abolition (Critical Resistance-INCITE, 2003; Davis, 2003; Davis et al., 2022; Paynter et al., 2022). A feminist abolitionist lens critically assesses how prisons produce harm, considers alternative approaches to prisons to address social harms, and rejects the conclusion that prisons can resolve gendered violence.
Data Acquisition
In 2018, we submitted an Access to Information and Privacy (ATIP) request to CSC for information regarding participation in the federal MCP from 2001 to the present (Set A and B). In July 2020, we submitted an ATIP request for information about MCP participants during the first wave of the COVID 19 pandemic (Set C). In late 2019, we received two data sets stemming from our first request: Set A included data from 2000–2015, and Set B from 2012–2019. We communicated with representatives from CSC's ATIP department by email and telephone to clarify information and terms used in the data set. We received Set C in September 2020. This data set includes April to August 10, 2020. We performed basic descriptive statistical analysis on the available data in each set using Excel.
Description of Data Sets
Set A was reportedly inputted by hand. Set B was drawn from the Offender Management System (OMS), a computerized case file management system used by CSC, on January 27, 2019. Set A and B overlapped for 2012–2015. There were discrepancies between the sets for years 2013–2015. We removed entries for years 2013–2015 from data Set A. The last year of data for Set B (2019) was only for January. Only data Set A tracks whether a participant used the program on a part-time or full-time basis. Only set B has information about participant Indigenous or non-Indigenous identity, and the amount of time in custody.
Results
Number of Participants
Both Set A and Set B have 67 participants in each. However, because of the overlap for 2012, there were 133 participants in the MCP program in total over 19 years. When Set B was pulled on January 27, 2019, there were 21 people flagged as active participants, of which five were still incarcerated. The others were on parole or had been released at their statutory release date.
In Set A, Joliette (QC) had the most participants (31) and Edmonton (AB) the least (1). In Set B, Edmonton had the most (21) and Fraser Valley (BC) the least (6). To put this in perspective, we include the (current) maximum capacity for number of women that can be incarcerated in each facility in Table 1. Participation was higher in Set B than Set A in all institutions except Joliette and Okimah Ochi. Participation generally fell over the course of Set A and rose over the course of Set B. This is in keeping with Brennan’s (2014) analysis that eligibility became more restrictive in 2008, reducing participation, and then additional beds were created in 2014 to improve participation. In Set A, there was an average of 5.2 new participants per year, and in Set B, the mean was 9.6. There were more new participants in 2016 (22) than any other year (See Figure 1). In total over the 20 years, Joliette (40), OkimawOhci (27) and Edmonton Institution for Women (22) had the most participants.
Set C includes eight participants across four of the institutions: Nova (1), Joliette (2), Grand Valley (1) and Fraser Valley (4). There were no participants at Edmonton Institution for Women or Okimaw Ohci. Information was not provided regarding Buffalo Sage.

New admissions to MCP 2000-2018. Note: 2019 is excluded as incomplete results were provided for the year.
Length of Participation
Length of participation in Set A ranged from 0 to 1,513 days; in Set B, from 4 to 1201 days. The mean days of participation were 168.3 days for Set A and 241.8 days for Set B. In both Set A and Set B, six people participated for less than a week's time.
Only Set B included the amount of time in years that each participant spent incarcerated. Fifty-six of the 67 (84%) of Set B participants were no longer incarcerated on January 27, 2019. Of those who had been released, they had served an average of 1.1 years in custody.
Discussion
This article is part of a series of research papers pertaining to the sexual, reproductive and maternal health of people experiencing incarceration in prisons for women in Canada (Paynter, Jefferies, et al., 2020; Paynter, Bagg, et al., 2020; Paynter et al., 2019 and Paynter et al., 2021). In our analysis, we use a feminist abolitionist framework (Critical Resistance-INCITE, 2003; Davis, 2003; Davis et al., 2022; Parkes, 2021; Paynter et al., 2022) critical of the suggestion that bringing children into prisons can resolve the problems of mother-child separation caused by the increasing incarceration of women. Not only is MCP participation under-researched, but there is little known about how many children are affected by parental incarceration in Canada. Their numbers are not tracked by Statistics Canada, CSC, or The Office of the Correctional Investigator (OCI). A case review of written sentencing decisions across the country in 2016 found not one mention of the Best Interests of the Child (Canadian Friends Service Committee, 2018). There is a need to address these information gaps, beginning with counting the children who are with their mothers in Canadian prisons.
From 2005–2015, the number of new participants in the MCP was very low, between 1 and 6 per year across all federal institutions. As Fayter and Payne (2017) point out, the MCP became less accessible under Prime Minister Stephen Harper's Conservative federal government, in power from 2006–2015. For some institutions, such as Edmonton, there was only a single participant during that time. Participation overall increased in recent years, but it fell at Joliette and Okimaw Ohci in Set B when compared to Set A.
There is no connection between size of facility or security levels in the facility with numbers of participants in the MCP (See Table 1). Most facilities have minimum, medium, and maximum-security sections. Policies across the facilities do not differ; all are subject to the same federal Commissioner's Directives and the Corrections and Conditional Release Act (1992).
Unlike Set A, which tracked the children, Set B tracks the mothers. As a result, we do not know how many children participate in Set B, as a mother may have more than one child with her, such as twins or a toddler and an infant. It is not necessary for participants to be pregnant and give birth while federally incarcerated. Furthermore, because it is the mothers that are flagged as participants, they may be entered into the program while pregnant, before the child is born. This obscures whether the child gets to participate at all, as the child may be removed from the mother's care at birth.
Indigenous identity is based on self-report. Although more granular data about self-reported race is collected by CSC, it was not available through ATIP. Information about participants’ identity as Indigenous or Non-Indigenous is not available for Set A. In Set B, 32.8% of participants identified as Indigenous. The most recent statistics show that 50% of federally incarcerated women identify as Indigenous (OCI, 2021). Indigenous women are underrepresented in the MCP, in keeping with Miller’s (2017) analysis that racism and colonialism impact security level classification and program eligibility. Although the proportion of incarcerated women who identify as Indigenous has increased every year over the time period covered in Set B, the proportion of MCP participants who identified as Indigenous varied from as low as 0% in 2012 and 2014 to as high as 56% in 2017.
Although only Set A tracked full versus part time participation, and time “in the program”, the number of days the child stays with their mother is not tracked. Mothers participating part-time could have their children with them every weekend, or only a few times a year. CSC representatives explained that it is also possible a mother was flagged as participating if her child came for a visit once. This is perhaps the explanation for participants with less than a week of participation.
Only Set B included information about the percent of sentence served and type of release. From Set B, we could discern that, for those participants who were paroled, they had spent an average of 1.1 years in custody. This is enough time to seriously disrupt their lives and leads to the question as to what purpose is served by incarcerating this population at all.
In the middle of our analysis, in March 2020, researchers, volunteers and visitors were largely locked out of federal facilities due to COVID-19. Advocacy organizations and academics around the world called for decarceration measures to prevent COVID-19 infections in prisons. This had little impact at the federal level in Canada. Outbreaks occurred in federal prisons across the country, including a large outbreak at Joliette Institution for Women (CSC, 2021; Thomas, 2020). Early in the pandemic, it became apparent that pregnant people were at greater risk of contracting COVID-19 and that COVID-19 was associated with increased risk of adverse birth outcomes including prematurity (CDC, 2021). Pregnant people experiencing incarceration faced compounding risks, a context not only stressful for those inside the institutions but for their family members in community (Germano, 2020). Despite this, Set C shows 8 women had their children with them in prison during the first wave of COVID-19.
In the context of a global pandemic and frequent prison outbreaks, the appropriateness of incarcerating children with their mothers can be called into question. Goshin (2015) has identified longstanding concerns with MCPs as a response to maternal incarceration, including the invisibility of children within the larger prison infrastructure, that the Best Interests of the Child (United Nations, 1989) and carceral policies do not align, and the lack of evidence-based guidelines to direct best practices. Some suggested alternatives to MCPs include “electronic tagging” such as ankle bracelets (Castello, 2015); and house arrest, as is used in Italy and Brazil for pregnant women and mothers of children under the age of 12 (Law Library of Congress, 2015). These approaches continue the carceral functions of confinement and surveillance. Supportive community-based housing models may be an alternative (Goshin, 2015). In France, alternatives must always be considered before the incarceration of pregnant women or mothers of young children (Blanchard et al., 2018).
Limitations
There are several limitations to our findings and the MCP data. It is a requirement of the United Nations Minimum Standards for the Treatment of Women Prisoners (United Nations, 2011) that a comprehensive health history including numbers of children be collected on a woman's admission to custody. CSC does not collect this information or data regarding maternal health outcomes such as gestational complications, breastfeeding, or peripartum depression. These data gaps impede any evaluation of the MCP, and any assessments of whether it improves or worsens these outcomes.
CSC representatives expressed concerns about the accuracy of the data in both Set A and B. However, there are no alternative sources for this information. The Canadian Association of Elizabeth Fry Societies, the national advocacy organization for people incarceratedin federal prisons for women, does not keep track. The OCI relies on data provided by CSC for its reports. Despite its limitations, information directly from CSC is the best available. It is unclear why our data set includes more participants than were reported by the OCI in 2019, as the OCI receives data for its reports from CSC. This discrepancy speaks to the general lack of attention to clear, consistent, reporting about this program.
Because the data are collected by flagging the day each new participant enters the program, there is no accurate picture of how many participants participated per year, but rather how many new people participated per year. Of the 67 entries in Set B, we found three instances where two participants had the exact same amount of time served in custody, to the third decimal point. While it is possible two unique individuals served such exactly similar amounts of time, it is unlikely. It is more likely that in these three instances one individual was entered twice in the system. For instance, if they were released and then were re-incarcerated, or moved from one institution to another, this would be captured as a new flagged start time.
Conclusion
Participation in the MCP in Canada over the past two decades was strikingly infrequent, averaging approximately 7 new participants in total each year. Despite the staggering increase in the numbers of federally incarcerated women over this time, MCP participation did not similarly increase. A disproportionately low number of Indigenous women participate in the MCP. Lack of high-quality data collection about participants, both mothers and children, hinders evaluation of the program. Considering the low numbers of participants, inadequacy of data for monitoring and evaluation of the MCP, not to mention the impact of COVID-19 on prisons across the country and around the world, there is an urgent need to re-evaluate the role of MCP and to consider investment in community-based alternatives.
Footnotes
Acknowledgements
Paynter would like to thank Clare Heggie and Sydney Keefe for research assistance. Paynter received doctoral funding from the Pierre Elliot Trudeau Foundation, the Canadian Institutes of Health Research Banting-Best Doctoral Scholarship, The Killam Laureate pre-doctoral funding program, the Canadian Nurses Foundation, the IWK Health Centre, and Dalhousie University.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The first author served as a volunteer with a prisoner rights organization.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article
