ORIGINAL RESEARCH: EMPIRICAL RESEARCH – QUANTITATIVE
Mothers’ group participation: associations with social capital, social support and mental well-being Cecily Strange, Alexandra Bremner, Colleen Fisher, Peter Howat & Lisa Wood Accepted for publication 3 August 2015
Correspondence to C. Strange: e-mail:
[email protected] The work was undertaken at the School of Population Health, The University of Western Australia Cecily Strange RN PhD School of Population Health, The University of Western Australia, Perth, Western Australia, Australia Alexandra Bremner PhD Assistant Professor in Biostatistics School of Population Health, The University of Western Australia, Perth, Western Australia, Australia Colleen Fisher PhD Professor/Head of School School of Population Health, The University of Western Australia, Perth, Western Australia, Australia Peter Howat PhD Professor School of Public Health, Curtin University, Perth, Western Australia, Australia Lisa Wood PhD Associate Professor School of Population Health, The University of Western Australia, Perth, Western Australia, Australia
STRANGE C., BREMNER A., FISHER C., HOWAT P. & WOOD L. (2016) Mothers’ group participation: associations with social capital, social support and mental well-being. Journal of Advanced Nursing 72(1), 85–98. doi: 10.1111/jan.12809
Abstract Aim. To investigate the relationships between participation in mothers’ groups and social capital, social support and mental well-being measures for mothers whose oldest child was 0-5 years. Background. Evaluations of facilitated mothers’ groups have found positive benefits for information sharing and support. Mothers’ groups often continue as parent-led groups; however, little is known about the potential benefits of ongoing participation compared with non-participation. Design. Cross-sectional survey. Methods. Data were collected through a survey from March 2013–January 2014 in Perth, Western Australia. The data from a subgroup of mothers (N = 313) whose oldest child was 0-5 years of age were analysed using multivariable regression. Participation in mothers’ groups in the previous 12 months was investigated for associations with social capital {Neighbourhood Cohesion Index (NCI); Families, Social Capital and Citizenship Survey (FSCCS) and Reciprocity}; social support {Medical Outcomes Study-Social Support Survey (MOS-SSS) and Parent Support Outside Home Scale (PSOHS)}; and mental well-being {Warwick Edinburgh Mental Well-Being Scale (WEMWBS)}. Participation was measured as three groups – locally, outside area of residence and non-participation. Results. Mothers who participated in mothers’ groups locally scored significantly higher than those who had not participated in mothers’ group for ‘social capital’ (NCI, FSCCS, Reciprocity), ‘social support’ (MOS-SSS, PSOHS) and ‘mental wellbeing’ (WEMWBS). Mothers who participated in mothers’ group outside the area scored significantly higher than those who had not participated in mothers’ groups for one measure of ‘social support’ (PSOHS). Conclusions. Participation in mothers’ group locally may provide support and social capital benefits for mothers of children aged 0-5 years, which may influence mental well-being. Keywords: community, maternity nursing, mental health, nurse roles, nursing, sociology, women’s health
© 2015 John Wiley & Sons Ltd
85
C. Strange et al.
Why is this research needed? • Facilitated mothers’ groups are recognized as opportunities for mothers to build a supportive social network. • Mothers’ groups often continue as parent-led groups, how-
tigate the relationships between participation in mothers’ groups, facilitated and/or parent-led, within the previous 12 months and social capital, social support and mental well-being measures for mothers whose oldest child was 05 years, compared with non-participation.
ever, little is known about the potential benefits from ongoing participation compared with non-participation.
What are the key findings? • Mothers who participated in mothers’ groups locally scored significantly higher than those who had not participated in mothers’ group for social capital (NCI, FSCCS, Reciprocity), social support (MOS-SSS, PSOHS) and mental well-being (WEMWBS). • Mothers who participated in mothers’ group outside the area scored significantly higher than those who had not participated in mothers’ groups for one measure of social support (PSOHS).
Implications for research, practice and policy • Parent-led groups commonly arise from facilitated ‘new mothers’ groups. Therefore, the findings in this study support the significant value of maternal and child health services engaged in initiating and facilitating mothers’ groups for new parents. • Local participation in a mothers’ group appears to have greater benefits than participation outside the area or nonparticipation. Mothers who do not participate in a mothers’ group locally may benefit from being linked into such groups. • Longitudinal research is needed to further investigate the associations found in this study.
Introduction ‘Mothers need others’ is an inferred truism argued by Hrdy (2011), an evolutionary socio-biologist. Mothers’ groups provide opportunities for new parents to build supportive networks and community connectedness and can be vital for mothers who would otherwise have a shortfall of social support (Strange et al. 2014a,b). Evaluations of mothers’ groups tend to focus on the period when they are facilitated (Fielden & Gallagher 2008, Guest & Keatinge 2009) and less is known about the longer term benefits from ongoing participation in parent-led mothers’ groups compared with non-participation (Scott et al. 2001, Strange et al. 2014b). There is a dearth of published quantitative investigation of the potential benefits of mothers’ group participation (facilitated or parent-led). The purpose of this study was to inves-
86
Background Mothers’ groups are facilitated in many countries in a variety of forms. The instigators and facilitators are often child health nurses (similar to health visitors in the UK), or hospital and parenting organizations supporting new parents. Common purposes of facilitated groups include providing information, support and fostering interaction for: new parents (Kruske et al. 2004, Guest & Keatinge 2009, Barnes et al. 2010, Schmied et al. 2014), all parents with infants (Kearney & Fulbrook 2012, Hj€ almhult et al. 2014) and for parents with specific needs such as teenage mums (Morrison et al. 2014). Facilitated groups also provide opportunities for parents to develop new social networks (Fielden & Gallagher 2008), preferably in the participant’s local community (Hj€ almhult et al. 2014). After the facilitation period of commonly 4-6 weeks, mothers are encouraged to continue meeting as a parent-led mothers’ group to provide ongoing peer support (Fielden & Gallagher 2008, Strange et al. 2014b). While a social support network is a reported benefit derived from ongoing participation in mothers’ groups (Scott et al. 2001), the evidence is scant. A recent qualitative study (Strange et al. 2014b) found mothers’ groups can provide four potential ongoing benefits for mothers of young children: peer learning, supportive networks, friendship and community connectedness. In particular, peer learning and support appear to reduce parental anxiety as experiences and knowledge are shared and normalized with others experiencing similar things (Strange et al. 2014b). Intrinsically, such benefits may be increasingly important due to social changes in the parenting landscape. The landscape of family life in Australia, as in many high income countries, has undergone significant change (Bianchi & Milkie 2010, Holmes et al. 2012). As a result of immigration, one in four (26%) Australians were born overseas. In Western Australia, the site of this study, the proportion of overseas born is 30%. In comparison, the proportions of overseas born in some other OECD countries are: New Zealand 23%, Canada 21%, USA 13% and UK 13% (Organisation for Economic Co-operation and Development (OECD) 2013). Australian families are often mobile as indicated by social survey data where 42% of adults overall
© 2015 John Wiley & Sons Ltd
JAN: ORIGINAL RESEARCH: EMPIRICAL RESEARCH – QUANTITATIVE
and 72% of 25-34 year olds reported a house move within the last five years (Australian Bureau of Statistics (ABS) 2010a), commonly coinciding with starting a family (Qu et al. 2012, Clark 2013). Women are returning to the workforce earlier after starting a family (Baxter 2013) and staying in the workforce longer compared with 20 years ago (ABS 2010b). Similar mobility and work trends have been experienced in other OECD countries (Kulu & Milewski 2008, Bianchi & Milkie 2010, Fagan & Norman 2012). The corollary is, families are more fragmented and traditional social support systems and social capital may be less accessible than in the past (McMurray 2007, Moore 2008, Leigh 2010). Social capital is increasingly part of the lexicon relating to social environments and well-being in communities (Carpiano 2008, Berry & Welsh 2010, Giordano et al. 2013) and is a construct of relevance to this study. For the purposes of this paper, which has a focus on families with young children, we refer to Harpham et al.’s (2002, p. 106) definition: ‘Social capital refer(s) to the degree of connectedness and the quality and quantity of social relations in a given population’. Social capital is also described as consisting of structural components or ‘what people do’ such as participation in social networks and cognitive components or ‘what people feel’ such as reciprocity, trust and shared values (Harpham et al. 2002, p,.106). Aspects of social cohesiveness and social support are viewed as features and resources of social capital by several authors (Harpham 2008, Baum et al. 2009, Berry & Welsh 2010) and this is congruent with the authors’ views in this study. Associations between social capital and physical and mental health outcomes have been confirmed in several reviews (De Silva et al. 2005, Islam et al. 2006, Murayama et al. 2012, McPherson et al. 2014). However, there has been limited evidence on the links between social capital and parental well-being, although there has been recognition of the importance of social network development for parents with young children (Fielden & Gallagher 2008, Kritsotakis et al. 2013, Hj€almhult et al. 2014). Perceived social support has been found to be a protective factor for the mental health of new mothers (LeahyWarren et al. 2012, Eastwood et al. 2013), disadvantaged mothers of children aged 0-4 years with depressive symptoms (Mulvaney & Kendrick 2005) and for mental and physical well-being more broadly (Berkman & Glass 2000). Perceived social support is commonly viewed and measured as functional forms: information, instrumental, emotional and appraisal (House et al. 1988, Sherbourne & Stewart 1991, Harpham 2008). For example, instrumental social © 2015 John Wiley & Sons Ltd
Associations with mothers’ group participation
support for families with young children may include child minding or other practical help. For families with young children, social support may be drawn from family and others external to the home. For those with limited access to extended family, social support from others in the community may be crucial (Strange et al. 2014b). Studies of maternal mental health for mothers of infants commonly focus on postnatal depression, which has been inversely associated with informal social support and parental self-efficacy (Leahy-Warren et al. 2012). However, several authors (see for example Bech et al. 2003, Tennant et al. 2007, Weich et al. 2011) have argued that measuring mental well-being, which is sometimes referred to as positive mental health, captures a wider scope of psychological well-being rather than the absence of anxiety or depression symptoms. This view is further supported by Weich et al.’s (2011) study which found that mental wellbeing, while correlated with mental illness, was generally independent, as people with poor mental health could still experience mental well-being. Positive mental health or well-being has been found to be associated with neighbourhood connections and social cohesion (Ziersch et al. 2005, Berry & Welsh 2010), particularly when good relationships exist in the local community (Jones et al. 2014).
The study Aim The aim was to investigate the relationships between participation in mothers’ groups, facilitated and/or parent-led, within the previous 12 months and social capital, social support and mental well-being measures for mothers whose oldest child was 0-5 years, compared with nonparticipation.
Design The study design was a cross-sectional survey. The questionnaire was informed by formative qualitative findings (Strange et al. 2014b, 2015) and piloted with parents of young children (N = 73).
Participants and data collection The results reported here are a part of a larger mixed methods study investigating how families with young children build social capital and feel connected and supported in newer residential areas in Perth, Western Australia. The larger study included survey data collected from parents who 87
C. Strange et al.
had at least one child aged 0-5 years (N = 489). Data were collected principally through an online survey from March 2013–January 2014. Only the data from mothers whose oldest child was aged 0-5 years (N = 331) are reported here. The vast majority (96%) of Australian households with children under 15 years have the Internet (ABS 2014) and formative research-identified preference for an online survey. Ensuring the administration of the survey suits the target audience is congruent with authors (Lyberg et al. 2012) on survey data collection. The first page of the survey included information and a statement indicating that completion of the questionnaire was considered consent to participate in the study. Recruitment was targeted towards newer residential areas as they are commonly populated by families with young children (ABS 2012). Furthermore, the physical and social infrastructure in newer residential areas is emergent (Elton Consulting 2012), which may impact on the opportunities for social interaction with others in the local community (Baum & Palmer 2002), particularly for families with young children (Andrews et al. 2014, Strange et al. 2015). Three local government areas (LGAs) in Perth with residential growth were selected for recruitment of participants through random selection of childcare centres and playgroups in those LGAs. There was a low response rate, however, possibly due to several factors such as: parents too busy caring for young children, no eligibility to enter a draw, or survey fatigue. Therefore, recruitment was revised. Recruitment was expanded to include seven additional LGAs in Perth with residential growth areas. Subsequently, agencies in the 10 LGAs contacted for recruitment promotion included child health centres, local government early childhood and families services, primary schools, all playgroups and childcare centres, where known and a few early childhood agencies, parent support websites and businesses with Facebook promotion features. In addition, the first author attended local government family events with hardcopy surveys and postcards with the survey link. Recruitment promotions included: hardcopy posters and postcards, and emails and Facebook postings with graphic invitations to participate along with survey links.
Study variables Mothers’ group participation Participation in mothers’ groups was assessed by asking respondents to indicate whether they had participated in a mothers’ group locally or outside their area of residence within the previous 12 months. The survey question did 88
not specify frequency of participation or whether the mothers’ group was facilitated or parent-led and, therefore, participation responses may include either or both forms of mothers’ groups and a range of participation frequency. Demographic variables Demographic variables included: parent age, number of children, relationship status, place of birth, education and employment status. Residential factors included: home ownership, age of residential area, distance from Perth central business district (CBD) and length of residence. Social capital Three different measures were used to capture different components of social capital. The first, Buckner’s (1988) Neighbourhood Cohesion Index (NCI), also used by Wood et al. (2013), was used in this study as a measure of ‘cognitive’ social capital or ‘what people feel’ at a local level (Strange et al. 2015). Buckner’s NCI consists of 18 items and includes several statements that resonate with the different experiences of community (e.g. ‘I feel like I belong to this neighbourhood’ and ‘I believe my neighbours would help me in an emergency’). Reliability of the NCI using Cronbach’s coefficient alpha has ranged from 091 (Robinson & Wilkinson 1995)-095 (Buckner 1988). Secondly, four items were included from the Families, Social Capital and Citizenship Survey (FSCCS) (Stone & Hughes 2002). The sum of the scores for these four items was used as applied by the Longitudinal Study of Australian Children (LSAC) (Zubrick et al. 2008), as a measure of ‘community connectedness’. The LSAC composite score classified parents who fell in the lowest 20% as scoring low for community connectedness (Zubrick et al. 2008). One item is a variation in the commonly used social capital question for trust ‘Most people in this neighbourhood can be trusted?’ and three items relate to being locally informed, seeking information and a sense of identity with the neighbourhood. Thirdly, reciprocity which is also a common social capital metric (Harpham 2008) was measured as the number of reciprocal categories (Box 1) the respondents were locally involved in the previous 12 months. This included the respondent doing something for someone in the neighbourhood or someone doing something in the neighbourhood for the respondent Social support Two social support scales were used in this study as the second measure was modelled on the first. The first comprised a modified form of the Medical Outcomes Study Social © 2015 John Wiley & Sons Ltd
JAN: ORIGINAL RESEARCH: EMPIRICAL RESEARCH – QUANTITATIVE
Support Survey (MOS-SSS) (Sherbourne & Stewart 1991) (13 of the 18 items, as used in the RESIDE study) (GilesCorti et al. 2008). Although the MOS-SSS was originally designed for use with patients with chronic conditions, it has also been used more widely (Gjesfjeld et al. 2008). It focuses on perceived availability of ‘functional’ social support, including instrumental, informational and emotional support and companionship (Sherbourne & Stewart 1991). The MOS-SSS scale has good internal consistency with reported Cronbach alpha coefficients >091 (Sherbourne & Stewart 1991) and more recently 083-094 for abbreviated versions (Gjesfjeld et al. 2008). The MOS-SSS is not place specific, so perceived support can be from within and/or external to the home. The second social support scale, (the Parent Support Outside Home Scale (PSOHS), was modelled on the MOS-SSS (Sherbourne & Stewart 1991) and developed specifically for this study to explore perceptions of functional social support available from outside the home, with a focus on support applicable to families with young children. The development of the PSOHS was guided by formative qualitative investigation (Strange et al. 2014a,b, 2015) and pilot testing. Ten items were designed to reflect different types of possible functional support available and respondents were asked to indicate the frequency of occurrence on a Likert scale from 1 (none of the time) to 5 (all of the time). The question and items are listed in Box 2. Mental well-being Positive mental well-being was measured using the Warwick Edinburgh Mental Well Being Scale (WEMWBS) which has
Box 1 Reciprocity categories Below is a list of activities neighbours sometimes do for one another. In the past 12 months: Which (if any) have you done for a neighbour or someone living in your neighbourhood? [Column A] Which (if any) has a neighbour or someone living in your neighbourhood done for you? [Column B] Cared for a child Cared for a family member other than a child Looked after a house or garden or collected mail or bins while away Minded, fed or walked a pet Loaned household or garden items or tools Listened to problems Provided advice or information Helped with odd jobs Provided lifts or transport (e.g. to the shops or school) Other (please write)
© 2015 John Wiley & Sons Ltd
Associations with mothers’ group participation
been validated in several studies (Tennant et al. 2007, Maheswaran et al. 2012, Stewart-Brown 2013) and consists of 14 items. Positive mental health encompasses both hedonic (affect, happiness and life satisfaction) and eudaimonic (cognitive and psychological functioning such as interpersonal relationships and perceived competence) perspectives (Tennant et al. 2007, Weich et al. 2011). Positive statements include for example: ‘I’ve been feeling optimistic about the future’ and ‘I’ve been thinking clearly’. The WEMWBS has had good internal consistency with Cronbach’s alpha scores between 083-091 (Tennant et al. 2007).
Ethical considerations This study was approved by the Human Research Ethics Committee at The University of Western Australia: RA/4/1/ 4958.
Data analysis Multivariable linear regression analyses were used to investigate associations between participation in mothers’ groups and each of the dependent variables (NCI, FSCCS Reciprocity, MOS-SSS, PSOHS and WEMWBS) with adjustment for demographic variables (number of children, relationship status, born overseas, education, employment, home ownership, age of residential area, distance from CBD and length of residence). Non-normal distributions (Reciprocity, MOS-
Box 2 Parent Support Outside Home Scale (PSOHS) items Parents of young children sometimes look to others from outside the home for different types of support. How often is each of the following kinds of support available, from outside the home, to you as a parent when you need it? Someone to give you good advice about your child’s behaviour e.g. feeding Someone to help in a practical way with your children e.g. child minding or care Someone to help in a practical way if your child or other family member is unwell Someone to help in a practical way if you are unwell Someone to share the joys of parenting with Someone to share the concerns of parenting with Someone to learn with about parenting by sharing experiences Someone to give you information or direct you to information on parenting Someone who makes you feel like you are doing a good job as a parent Someone for you and your children together to spend time with
89
C. Strange et al.
SSS and PSOHS) were logged to normalize residual plots as required. SPSS version 21 (IBM 2012) was used for data analysis and the significance level was set at 005.
Table 1 Demographic characteristics by mothers’ group (MG) participation and non- participation.
Validity and reliability Validated survey tools were used with the exception of the ten item scale PSOHS. To ascertain validity of this scale, a pilot of the PSOHS was tested (T1) with a convenience sample (N = 73) and then retested (T2). The pilot intraclass correlation coefficients (ICC) for the PSOHS of 042-067 fell within the accepted range (Dawson & Trapp 2004). The internal consistency reliability for PSOHS was assessed using Cronbach alpha coefficients (pilot and main questionnaire) and ranged from 093-095 and inter-item correlations ranged from 053-092, which indicated good internal consistency (Cronbach 1951). Principal axis factoring analysis was undertaken (main survey) to explore the structure of the PSOHS for subsets that accounted for common variance. Two factors with eigenvalues exceeding one were identified: 1) instrumental support; and 2) information, emotional and companionship support. There were no indications that any of the items of the PSOHS should be removed. The PSOHS and MOS-SSS were strongly correlated (r = 071).
Results Table 1 presents the demographic characteristics (N = 331) of participants in mothers’ groups within the previous 12 months (locally and outside the area) and non-participants. As indicated, 659% participated in mothers’ groups. Chi-square results indicated mothers’ group participants were significantly more likely to have only one child (P = 0025), have home ownership (P < 0001) and to have lived in their current suburb three years or longer (P = 0022) compared with those who did not participate in mothers’ groups within the last 12 months. Regression analyses was undertaken of data from the subgroup of mothers (N = 313) who either participated in mothers’ groups locally (N = 149) or outside the area (N = 51), or were non-participants (N = 113). Data from those that participated in a mothers’ group both locally and outside the area (N = 18) were not included in the regression analysis due to small group size. As age of parents was not significantly associated with any of the dependent variables, it was not included as an adjustment variable. Higher scores indicate greater levels of reported social capital (NCI, FSCCS and Reciprocity), social support (MOS-SSS and PSOHS) and mental well-being (WEMWBS). 90
Mothers with oldest child 0-5 years of age (n = 331)
Participated in MG locally or outside the area in the last 12 months (n = 218) 659%
Did not participate in MG at all in the last 12 months (n = 113) 341%
N
N
%
Age of parents (n = 330) 18-29 years 63 288 30-39 138 638 40 and older 16 74 Number of children in family 1 118 539 2 89 406 3 or more 11 55 Relationship status Married or living 208 954 with partner Single, separated, 10 46 divorced or widowed Place of birth (n = 330) Australia 151 696 Overseas 66 304 Education attained (n = 330) High school year 33 156
