Sage Journals: Discover world-class research

Abstract

French

Objective:

The objective of this study is to examine suicide attempts and completions among mothers who had a child taken into care by child protection services (CPS). These mothers were compared with their biological sisters who did not have a child taken into care and with mothers who received services from CPS but did not have a child taken into care.

Methods:

A retrospective cohort of mothers whose first child was born in Manitoba, Canada, between April 1, 1992, and March 31, 2015, is used. Rates among discordant biological sisters (1872 families) were compared using fixed-effects Poisson regression models, and mothers involved with CPS (children in care [n = 1872] and received services [n = 9590]) were compared using a Poisson regression model.

Results:

Compared with their biological sisters and mothers who received services, the adjusted incidence rate ratio (aIRR) of death by suicide was greater among mothers whose child was taken into care by CPS (aIRR = 4.46 [95% confidence interval (CI), 1.39-14.33] and ARR = 3.45 [95% CI, 1.61-7.40], respectively). Incidence rates of suicide attempts were higher among mothers with a child taken into care compared with their sisters (aIRR = 2.15; 95% CI, 1.40-3.30) and mothers receiving services (aIRR = 2.82; 95% CI, 2.03-3.92).

Conclusions:

Mothers who had a child taken into care had significantly higher rates of suicide attempts and completions. When children are taken into care, physician and social workers should inquire about maternal suicidal behaviour and provide appropriate mental health.

Keywords

cohort study epidemiology maternal health suicide child protection services linkable administrative data siblings

The impact of being placed in care of child protection services (CPS) on children has been examined extensively. These children have worse health and social outcomes than those of children in the general population.^1

–5 Several population-based studies have also identified these children as being at higher risk for suicide and suicide attempts.^5
–7 Less known are the outcomes of mothers whose children are placed in care by CPS. While mothers with mental illness have significantly greater involvement with CPS, women have reported that they developed mental health conditions, or their existing mental health conditions worsened, after custody loss.^8,9 Being a mother has been identified as being protective against suicide; however, experiencing the death of a child increases this risk, particularly if the child is young.¹⁰ Parents who lose custody of their children have been shown to experience this situation as a ‘living death’.¹¹ The purpose of this study is to examine whether mothers who lost custody of their children due to involvement with CPS have higher rates of suicide attempts and completions than mothers not separated from their children.

Previous observational studies investigating rates of suicide in at-risk populations have controlled for individual mental health, socioeconomic status, and various demographic variables, as these factors have been linked to suicide and suicide attempts.^12,13 Mothers whose children are taken into care often have very different life circumstances than mothers in the general population. To account for some of these differences, we used to comparison groups. The first comparison is between mothers who had a child taken into care and their biological sisters who did not have a child taken into care. This was done because suicide attempts and completions often cluster within families due to shared genetic and environmental factors, and family fixed-effects models of discordant siblings account for stable family characteristics.^14
–16 The second comparison was between mothers who had a child taken into care and mothers who received services from CPS but were not separated from their child or children. This comparison was made as the instability that resulted in involvement with CPS may also contribute to differences in suicide attempts and completions. While the comparison groups are imperfect, both capture important similarities.¹⁷

Methods

Setting

Manitoba is a central Canadian province with approximately 1.2 million residents at the time of the 2011 census.¹⁸ Manitoba residents receive universal health care coverage. While Manitoba is representative of Canada in many aspects, ranking in the middle for several health indicators and slightly below average on education indicators, rates of children placed in out-of-home care are among the highest in the country.^19,20 With approximately 3% of children in care, Manitoba has the highest rates of children in care among Canadian provinces and one of the highest rates in the world.²¹ Several Canadian provinces and territories have similarly high rates, but provinces with rates lower than those seen in Manitoba still have higher rates than many other countries.²² Suicide is a leading cause of death among young Canadians, with a rate of 4.8 per 100,000 in the female population in 2012.^23,24 This is above the rate of 2.6 in the United Kingdom and below the rate of 5.6 seen in the United States.²⁴ Manitoba ranks in the middle of suicide rates among Canadian provinces and territories.²⁵

Data

This study uses linkable administrative data found in the Population Research Data Repository housed at the Manitoba Centre for Health Policy. This Repository contains province-wide, routinely collected individual data over time and across geographic space for each provincial resident.²⁶ Births, deaths, arrival, and departure dates are identified for each resident, with follow-up comparable to that of the largest cohort studies based on primary data.²⁷ This research links data from the population registry with individual-level information from physician claims, hospital discharge abstracts, vital statistics, Child and Family Services (CFS) case reports, and the Canadian Census. An encrypted personal health number linked these de-identified data sets. Information on linkage methods, confidentiality/ privacy, and validity can be found elsewhere.^28,29 Children are linked to mothers using hospital birth record information; mothers are noted in essentially all cases.³⁰

Cohort Formation

Individuals included in this study were drawn from the whole Manitoba population of women whose first child was born in Manitoba between April 1, 1992, and March 31, 2015. The cohort was divided into 3 groups. Group 1 included all women having at least 1 child taken into care before March 31, 2015. For mothers having multiple children removed from their home, the oldest child was selected as the index child; the index date is the date the index child was first taken into care. Mothers in group 1 may have sisters who also had a child taken into care; to ensure independence, 1 sister is randomly selected in group 1. Group 2 included biological sisters of mothers in group 1 who did not have a child taken into care. The index date for these mothers is the date that their oldest child turns the age their sister's child was when taken into care. Again, 1 sister per family was randomly selected when families had more than 1 sister in group 2. Group 3 is a group of mothers without a sister in the cohort; these women received other protection or support services from CPS but did not have a child taken into care. Other protection or support services are provided to resolve family matters, with the children not removed from the home; these services include counselling, guidance, support, education, and emergency shelter services.^31,32 The index date for mothers in group 3 is the date they first began receiving services from CPS. Mothers not living in Manitoba for at least 2 years before the index date, or those for whom the index date was after March 31, 2015, were excluded. Figure 1 diagrams the cohort formation.

Figure 1.

Cohort selection.

Variables

Outcomes

For each mother, we examined suicide attempts and completions from the index date up to December 31, 2015 (see Table 1 for definitions). Both variables were defined using administrative data, with definitions based on previous work using these data.

Table 1.

Definition of Variables.

Outcomes	Definition
Suicide completion after index date	Death due to suicide as recorded in the vital statistics database any time between the index date and December 31, 2015, including ICD-10 codes X40-X42, X46, X47, X60-X84.⁵
Suicide attempt after index date	Hospitalization between the index date and December 31, 2015, with ICD-9-CA diagnosis code E950 to E959 (before April 1, 2004) or ICD-10-CA diagnosis codes X60 to X84 (after April 1, 2004).³³
Covariates
Suicide attempt before index date	Hospitalization in the 2 years before the index date with ICD-9-CA diagnosis codes E950 to E959 (before April 1, 2004) or ICD-10-CA diagnosis codes X60 to X84 (after April 1, 2004).³³
Substance use diagnosis before index date	Hospitalizations in the 2 years before the index date with ICD-9-CM diagnosis codes 291, 292, and 303 to 305 (before April 1, 2004) or ICD-10-CA diagnosis codes F10 to F19 and F55 (after April 1, 2004), or a physician visit in the 4 years before the index date with ICD-9-CM diagnosis codes 291, 292, and 303 to 305.³⁴
Mood and anxiety disorder diagnosis before index date	In the 2 years before the index date, the mother has (1) a hospitalization with a diagnosis of ICD-9-CM codes 296.1 to 296.8, 300.0, 300.2 to 300.4, 300.7, 309, and 311 (before April 1, 2004) or ICD-10-CA codes F31, F32, F33, F34.1, F38.0, F38.1, F40, F41.0 to F41.3, F41.8, F41.9, F42, F43.1, F43.2, F43.8, F45.2, F53.0, and F93.0 (after April 1, 2004); (2) a physician visit with a diagnosis of ICD-9-CM codes 296 and 311; or (3) at least 3 physician visits with diagnosis of ICD-9-CM code 300 or 309.^35,36
Socioeconomic status of neighbourhood	The socioeconomic status of the neighbourhood of birth was measured by the Socioeconomic Factor Index score (higher score corresponds with lower socioeconomic status) generated using small census dissemination areas and classified by urban/rural status.³⁷ This score was standardized at the provincial level (before removing exclusions and missing values).

Covariates

Potential confounders were selected based on clinical relevance; among the best predictors of suicide or suicide attempts are previous suicide attempts and mental illness.³⁸ Each model adjusted for a series of maternal mental health conditions in the 2 years before the index date: suicide attempts (yes/no), number of substance use diagnoses (0, 1, 2+), and number of mood and anxiety disorder diagnoses (0, 1, 2+) (see Table 1 for definitions). We included the number of diagnoses in the 2 years before the index date to estimate the severity of the diagnosis. Socioeconomic status, mother’s age, and child’s age have also been identified as predictors of suicide attempts and completions.^39,40 Each model adjusted for socioeconomic status (see Table 1 for definition) and location of neighbourhood (urban, rural south, rural mid, rural north), age of mother, and age of child at the index date.

Ethics

This study was reviewed and approved by the Health Research Ethics Board at the University of Manitoba (#H2016:182) and the Health Information Privacy Commission at Manitoba Health, Seniors and Active Living (#2016/2017-09). This study used de-identified administrative data files, which did not require informed consent from participants.

Statistical Analysis

Two analyses were conducted. First, mothers having a child taken into care (group 1) were compared with their biological sisters who did not have a child taken into care (group 2). Adjusted incidence rate ratios (aIRRs) for each outcome were obtained from family fixed-effects Poisson regression models. This reduced the potential for confounding, as this approach helps account for shared familial characteristics, environment, and genetic predisposition.⁴¹ The second analysis compared mothers with a child taken into care with mothers who received other protection or support services from CPS (but did not have a child taken into care). Adjusted IRRs for each of the outcomes were obtained using Poisson regression models. For each analysis, the log of the total number of person-years at risk was included as an offset to ensure modeling of a rates as opposed to a count of events. Data management, programming, and analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC).

Results

Overall, 13,334 mothers were included in this study. Of these mothers, 1872 had a child taken into care, 1872 were biological sisters not having a child taken into care, and 9590 mothers received services from CPS but did not have a child taken into care. Mothers were followed from the index date until their coverage ended or December 31, 2015 (whichever came first), resulting in a total of 131,582 person-years of follow-up. Those in group 1 were followed for an average of 8.96 years, their sisters (group 2) were followed for an average of 8.38 years, and the mothers (group 3) receiving services were followed for an average of 10.34 years. Mothers whose children were taken into care were younger and lived in neighbourhoods of lower socioeconomic status at the index date. These mothers were also more likely to have had a suicide attempt and had more substance use diagnoses in the 2 years before the index date (Table 2). Almost all mothers (99%) had contact with a physician after the index date.

Table 2.

Characteristics of Mothers, Manitoba, Canada, 1992-2015.

Covariates	Group 1: Had a Child Taken into Care (n = 1872)	Did Not Have a Child Taken into Care
Covariates	Group 1: Had a Child Taken into Care (n = 1872)	Group 2: Biological Sisters (n = 1872)	Group 3: Received Services (n = 9590)
At the index date, mean (SD)
Age of child	3.97 (4.50)	4.19 (4.56)	3.42 (4.31)
Age of mother	24.82 (6.77)	25.33 (6.57)	27.18 (8.81)
Socioeconomic status of neighbourhood^a	1.18 (1.13)	1.13 (1.24)	0.37 (1.05)
Location of neighbourhood, n (%)
Urban	950 (50.75)	842 (44.98)	6296 (65.65)
Rural south	239 (12.77)	272 (14.53)	1410 (14.70)
Rural mid	296 (15.81)	318 (16.99)	1164 (12.14)
Rural north	387 (20.67)	440 (23.50)	720 (7.51)
Maternal diagnoses in the 2 years before the index date, n (%)
Suicide attempts	27 (1.44)	8 (0.43)	34 (0.35)
Number of mood and anxiety disorder diagnoses
0	1133 (60.52)	1426 (76.18)	6052 (63.11)
1	258 (13.78)	189 (10.10)	1326 (13.83)
2+	481 (25.69)	257 (13.73)	2212 (23.07)
Number of substance use disorder diagnoses
0	1548 (82.69)	1743 (93.11)	9039 (92.25)
1	178 (9.51)	85 (4.54)	355 (3.70)
2+	146 (7.80)	44 (2.35)	196 (2.04)

^aSocioeconomic status is measured by the Socioeconomic Factor Index (SEFI); higher values correspond with lower socioeconomic status.³⁷

Mothers whose children were taken into care had 286 additional suicide attempts and 76 suicide completions per 100,000 person-years compared with their sisters who did not have a child taken into care (Table 3). Compared with mothers who received services, mothers whose children were taken into care had 352 additional suicide attempts and 77 suicide completions per 100,000 person-years.

Table 3.

Rates (per 100,000 Person-Years) of Suicide Attempts and Completions, Manitoba, Canada, 1992-2015.

Outcome	Group 1: Had a Child Taken into Care	Did Not Have a Child Taken into Care
	Group 1: Had a Child Taken into Care	Group 2: Sisters		Group 3: Received Services
	Incidence Rate^a (95% CI)	Incidence Rate^a (95% CI)	Rate Difference^b (95% CI)	Incidence Rate^a (95% CI)	Rate Difference^b (95% CI)
Suicide attempt	470.9 (377.7-387.1)	184.8 (128.4-265.9)	286.1 (234.6-337.6)	118.0 (98.5-141.5)	352.9 (305.4-400.4)
Suicide completion	95.4 (58.4-155.7)	19.1 (6.2-59.3)	76.3 (24.8-127.8)	18.2 (11.4-28.8)	77.2 (29.7-124.7)

CI, confidence interval.

^aPer 100,000 person-years.

^bCompared with mothers in group 1; per 100,000 person-years.

Compared with sisters and mothers who received services, the incidence rate of suicide attempts and completions was significantly higher among mothers having a child taken into care by CPS (Table 4). The sister comparison helps control for shared familial and genetic confounding, whereas the comparison with mothers who received services controls for maternal instability requiring CPS involvement.

Table 4.

Unadjusted and Adjusted Incidence Rate Ratios of Suicide Attempts and Completions, Manitoba, Canada, 1992-2015.

Outcome	Child Taken into Care vs Sisters		Child Taken into Care vs Received Services
Outcome	Unadjusted IRR (95% CI)	Adjusted^a IRR (95% CI)	Unadjusted IRR (95% CI)	Adjusted^a IRR (95% CI)
Suicide attempt	2.55 (1.68-3.86)	2.15 (1.40-3.30)	3.99 (3.00-5.31)	2.82 (2.03-3.92)
Suicide completion	4.99 (1.55-16.02)	4.46 (1.39-14.33)	5.25 (2.68-10.30)	3.45 (1.61-7.40)

CI, confidence interval; IRR, incidence rate ratio.

^aAdjusted for all variables in Table 2.

Discussion

To our knowledge, this study is the first to examine suicide attempts and completions among mothers who had a child taken into care by CPS. Mothers who have a child placed in care have much higher rates of suicide attempts and completions than their biological sisters and mothers who received services, even when adjusting for many risk factors.

These results of our study are concordant with, and may be explained by, earlier findings indicating that mothers losing custody of their child to CPS have increased distress and psychiatric problems.⁹ Most mothers experience a great deal of pain when they lose custody of a child, which often leads to a loss of self-worth and confidence.^42,43 Regardless of their situation, mothers want to be seen as ‘good mothers’, and when a woman associates being a ‘good mother’ with being a ‘good person’, the involvement with child protective services will place in question her mothering and her identity.⁴⁴ Feelings of guilt, responsibility, shame, stigmatization, and loss of self-worth often associated with this type of custody loss may also contribute to the increased rates of suicide attempts and completions.^42,43,45

Mothers whose children are taken into care often have existing mental health conditions that contributed to the loss of custody of their children.⁴⁶ Mothers have reported the development and worsening of mental health conditions after separation from their child.^8,9 Strengthening resources and capabilities of mothers who are at risk of having a child taken into care may result in better outcomes for both mother and child.^47
–49 In situations where a child does need to be removed from the home, service providers ensure that the biological mothers are receiving adequate mental health supports. This could both prevent the deterioration of mental health in these mothers and ensure that the stresses associated with the separation do not become another barrier to reunification.

The use of the population-based repository at the Manitoba Centre for Health Policy offers significant strengths, including a large sample size, minimal attrition, and a large number of predictors over many years.⁵⁰ Study limitations concern unmeasured confounders and the measurement of variables. This study only captures physician-treated mental illness. Therefore, we may be missing women who experience mental illness but do not seek treatment or treatment from physicians, resulting in an underestimation of history of mental illness in our cohort.⁵¹ In addition, for both mood and anxiety disorders and substance use disorders before the index date, we do not know the severity of the disease, its duration, or resulting disability. Suicide is often undercoded in administrative data, as physicians may list underlying mental illness as the diagnosis or not accurately chart the occurrence of suicidal behaviour.⁵² These data have low sensitivity in tracking the prevalence of suicide outcomes; using hospital and physician claims data to identify suicidal outcomes in patients may undercount one-half to two-thirds of these outcomes.⁵² Only women who received physician treatment owing to a suicide attempt or whose death has been classified as a suicide were included. Our definition of suicide using vital statistics files excludes deaths related to motor or other vehicle collisions, which could potentially be suicide-related deaths (particularly in rural areas). These limitations may result in the underestimation of the burden of suicide among our study population. As in many observational studies, the results of this study may be biased by unmeasured confounders. Variables that have been linked with suicide attempts and completion and were not accounted for in this study include maternal education, domestic violence, maternal history of abuse and neglect, and Indigenous status.^7,53,54 In Manitoba, Indigenous individuals are more likely to be involved with CPS and also have higher rates of suicide and suicide attempts.^20,55 While this confounder was not adjusted for in the analysis comparing mothers whose children were taken with mothers who received services, the discordant sibling analysis takes Indigenous status into account (because it is a stable family characteristic). Methodologically, this work builds on efforts to incorporate family variables into cohort studies.⁵⁶ The setting of this research—Manitoba, Canada—has a very high rate of children in care.²¹ To ensure generalizability, this study needs to be replicated in other settings with different policies around child welfare.

Conclusion

Women who have children taken into care have increased psychological distress and are at greater risk for suicide and suicide attempts. For mothers whose children are taken into care, greater mental health supports should be provided to minimize the impact of this event and to address any underlying mental health issues contributing to the child being taken into care.

Footnotes

Acknowledgement

Data used in this study are from the Population Research Data Repository housed at the Manitoba Centre for Health Policy, University of Manitoba and were derived from data provided by Manitoba Health, Seniors and Active Living, Manitoba Families, and Vital Statistics under project 2016/2017-09. The results and conclusions are those of the authors and no official endorsement by the Manitoba Centre for Health Policy, Manitoba Health, Seniors and Active Living, or other data providers is intended or should be inferred.

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.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Preparation of this manuscript was supported by a Social Sciences and Humanities Research Council of Canada Joseph-Armand Bombardier Canada Doctoral Scholarship, a Graduate Enhancement of Tri-council Stipend, and a Women’s Health Research Foundation of Canada Full Time Scholarship.

ORCID iD

Elizabeth Wall-Wieler

References

Clausen

Landsverk

Ganger

. Mental health problems of children in foster care. J Child Fam Stud. 1998;7(3):283–296.

Leslie

Hulburt

James

. Relationship between entry into child welfare and mental health services use. Psychiatr Serv. 2014;56(8):981–987.

Putnam-Hornstein

Needell

. Predictors of protective services contact between birth and age five: an examination of California’s 2002 birth cohort. Child Youth Serv Rev. 2014;33:1337–1344.

Cosgrove

Frost

Chown

. Strengthening health outcomes for foster care children. Madison (WI): University of Wisconsin-Madison; 2013.

Katz

Singal

. Suicide and suicide attempts in children and adolescents in the child welfare system. CMAJ. 2011;183(17):1977–1981.

Hjern

Vinnerljung

Lindblad

. Avoidable mortality among child welfare recipients and intercountry adoptees: a national cohort study. J Epidemiol Community Health. 2004;58(5):412–417.

Vinnerljung

Hjern

Lindblad

. Suicide attempts and severe psychiatric morbidity among former child welfare clients—a national cohort study. J Child Psychol Psychiatry. 2006;47(7):723–733.

Hollingsworth

. Child custody loss among women with persistent severe mental illness. Soc Work Res. 2004;28(4):199–209.

Kenny

Barrington

Green

. “I felt for a long time like everything beautiful in me had been taken out”: women’s suffering, remembering, and survival following the loss of child custody. Int J Drug Policy. 2015;26:1158–1166.

10.

Qin

Mortensen

. The impact of parental status on the risk of completed suicide. Arch Gen Psychiatry. 2003;60(8):797–802.

11.

Zeman

. Etiology of loss among parents falsely accused of abuse or neglect. J Loss Trauma. 2004;10(1):19–31.

12.

Blakely

. Unemployment and suicide: evidence for a causal association? J Epidemiol Community Health. 2003;57(8):594–600.

13.

Agerbo

Mortensen

Erickson

. Risk of suicide in relation to income level in people admitted to hospital with mental illness: nested case-control study. BMJ. 2001;322(7282):334–335.

14.

Agerbo

Qin

Mortensen

. Psychiatric illness, socioeconomic status, and marital status in people committing suicide: a matched case-sibling-control study. J Epidemiol Community Health. 2006;60:776–781.

15.

Kim

. Familial aggregation of suicidal behavior: a family study of male suicide completers from the general population. Am J Psychiatry. 2005;162(5):1017–1019.

16.

Qin

. Suicide risk in relation to family history of completed suicide and psychiatric disorders: a nested case-control study based on longitudinal registers. Lancet. 2002;360(9340):1126–1130.

17.

D’Onofrio

Lahey

Turkheimer

. Critical need for family-based, quasi-experimental designs in integrating genetic and social science research. Am J Public Health. 2013;103(1 suppl):46–56.

18.

Statistics Canada. Focus on Geography Series, 2011. 2014. Available from: http://www12.statcan.gc.ca/census-recensement/2011/as-sa/fogs-spg/Facts-pr-eng.cfm?&GK=PR&GC=46. Accessed January 4, 2016.

19.

O’Grady

Deussing

Scerbina

. Measuring up: Canadian results of the OECD PISA Study. Toronto (ON): 2016. Available from: http://cmec.ca/Publications/Lists/Publications/Attachments/365/Book_PISA2015_EN_Dec5.pdf. Accessed January 3, 2017.

20.

Brownell

Chartier

. The educational outcomes of children in care in Manitoba. Winnipeg (MB); 2015. Available from: http://mchp-appserv.cpe.umanitoba.ca/reference/CIC_report_web.pdf. Accessed September 24, 2016.

21.

Gilbert

Fluke

O’Donnell

. Child maltreatment: variation in trends and policies in six developed countries. Lancet. 2012;379(9817):758–772.

22.

Thoburn

. Globilisation and child welfare: some lessons from a cross-national study of children in out-of-home care. Norwich (England); 2007. Available from: http://www.uea.ac.uk/swk/research/publications/monographs/welcome.htm. Accessed April 27, 2017.

23.

Navaneelan

. Suicide rates: an overview. Stat Canada. 2015 [cited 2017 Feb 17]. Available from: http://www.statcan.gc.ca/pub/82-624-x/2012001/article/11696-eng.htm

24.

World Health Organization. Suicide rates data by country. 2016 [cited 2017 Feb 17]. Available from: http://apps.who.int/gho/data/node.main.MHSUICIDE?

25.

The Conference Board of Canada. Suicides. 2015. Available from http://www.conferenceboard.ca/hcp/provincial/health/suicide.aspx. Accessed February 19, 2017.

26.

Nickel

Chateau

Martens

. Data resource profile: pathways to health and social equity for children (PATHS Equity for Children). Int J Epidemiol. 2014;43(5):1438–1449.

27.

Power

Kuh

Morton

. From developmental origins of adult disease to life course research on adult disease and aging: insights from birth cohort studies. Annu Rev Public Health. 2013;34:7–28.

28.

Roos

Gupta

Soodeen

. Data quality in an information-rich environment: Canada as an example. Can J Aging. 2005;24(suppl 1):153–170.

29.

Roos

Nicol

. A research registry: uses, development, and accuracy. J Clin Epidemiol. 1999;52(1):39–47.

30.

Currie

Stabile

Manivong

. Child health and young adult outcomes. J Hum Resour. 2010;45(3):517–548.

31.

Family Services and Housing. Child protection services. Gov Manitoba. 2009. Available from: www.gov.mb.ca/fs/cfsmanual/pubs/pdf/1.3.1_enp.pdf. Accessed February 2, 2016.

32.

Manitoba Centre for Health Policy. Term: protection services. 2007 [cited 2016 Dec 15]. Available from: http://mchp-appserv.cpe.umanitoba.ca/viewDefinition.php?definitionID=103841

33.

Manitoba Centre for Health Policy. Concept: suicide and attempted suicide (Intentional Self Inflicted Injury). 2016 [cited 2016 Oct 26]. Available from: http://mchp-appserv.cpe.umanitoba.ca/viewConcept.php?conceptID=1183

34.

Chartier

Brownell

MacWilliam

. The mental health of Manitoba’s children. Winnipeg (MB): Manitoba Centre for Health Policy; 2016. Available from http://www.mchp-appserv.cpe.umanitoba.ca/reference/MHKids_web_report.pdf

35.

Manitoba Centre for Health Policy. Concept: mood and anxiety disorders—measuring prevalence. 2015 [cited 2016 Jan 30]. Available from: http://mchp-appserv.cpe.umanitoba.ca/viewConcept.php?conceptID=1391

36.

Kisely

Lin

Gilbert

. Use of administrative data for the surveillance of mood and anxiety disorders. Aust N Z J Psychiatry. 2009;43(12):1118–1125.

37.

Chateau

Metge

Prior

. Learning from the census: the socio-economic factor index (SEFI) and health outcomes in Manitoba. Can J Public Health. 2012;103(suppl 2):S23–S27.

38.

Wenzel

Berchick

Tenhave

. Predictors of suicide relative to other deaths in patients with suicide attempts and suicide ideation: a 30-year prospective study. J Affect Disord. 2011;132(3):375–382.

39.

Rehkopf

Buka

. The association between suicide and the socio-economic characteristics of geographical areas: a systematic review. Psychol Med. 2006;36(2):145–157.

40.

McGirr

Renaud

Bureau

. Impulsive-aggressive behaviours and completed suicide across the life cycle: a predisposition for younger age of suicide. Psychol Med. 2008;38(3):407–417.

41.

Donovan

Susser

. Commentary: advent of sibling designs. Int J Epidemiol. 2011;40(2):345–349.

42.

Hollingsworth

. Birth mothers whose parental rights are terminated: implications for services. In: Mallon

Hess

, editors. Child welfare for the twenty-first century: a handbook of practices, policies, and programs. New York (NY): Columbia University Press; 2005. p. 469–481.

43.

Nixon

Radtke

Tutty

. “Every day it takes a piece of you away”: experiences of grief and loss among abused mothers involved with child protective services. J Public Child Welf. 2013;7(2):172–193.

44.

Wells

. A narrative analysis of one mother’s story of child custody loss and regain. Child Youth Serv Rev. 2011;33:439–447.

45.

Lester

. The role of shame in suicide. Suicide Life Threat Behav. 1997;27(4):352–361.

46.

O’Donnell

Maclean

Sims

. Maternal mental health and risk of child protection involvement: mental health diagnoses associated with increased risk. J Epidemiol Community Health. 2015;69(12):1175–1183.

47.

Shonkoff

Fisher

. Rethinking evidence-based practice and two-generation programs to create the future of early childhood policy. Dev Psychopathol. 2013;25(402):1635–1653.

48.

Fisher

Frenkel

Noll

. Promoting healthy child development via a two-generation translational neuroscience framework: the filming interactions to nurture development video coaching program. Child Dev Perspect. 2016;10(4):251–256.

49.

Chartier

Brownell

Issac

. Is the families first home visiting program effective in reducing child maltreatment and improving child development? Child Maltreat. 2017;22(2):121–131.

50.

Roos

Walld

Burchill

. Linkable administrative files: family information and existing data. Longit Life Course Stud. 2017;8(3):264–280.

51.

Wang

Aguilar-Gaxiola

Alonso

. Use of mental health services for anxiety, mood, and substance disorders in 17 countries in the WHO world mental health surveys. Lancet. 2007;370(9590):841–850.

52.

Randall

Roos

Lix

. Emergency department and inpatient coding for self-harm and suicide attempts: validation using clinician assessment data. Int J Methods Pyschiatric Res. 2017;26(3).

53.

Olfson

Blanco

Wall

. National trends in suicide attempts among adults in the United States. JAMA Psychiatry. 2017 9 13. [Epub ahead of print]

54.

Vijayakumar

. Suicide in women. Indian J Psychiatry. 2015;57(2 suppl):S233–S238.

55.

The Aboriginal Healing Foundation. Suicide among aboriginal people in Canada. Ottawa (ON); 2007. Available from: http://www.ahf.ca/downloads/suicide.pdf. Accessed February 19, 2017.

56.

McKeague

Lumey

. Optimal design strategies for sibling studies with binary exposures. Int J Biostat. 2014;10(2):185–196.

Suicide Attempts and Completions among Mothers Whose Children Were Taken into Care by Child Protection Services: A Cohort Study Using Linkable Administrative Data

Abstract

Objective:

Methods:

Results:

Conclusions:

Keywords

Methods

Setting

Data

Cohort Formation

Variables

Outcomes

Covariates

Ethics

Statistical Analysis

Results

Discussion

Conclusion

Footnotes

Acknowledgement

Declaration of Conflicting Interests

Funding

ORCID iD

References