Women's perspectives on resilience and research on resilience in motherhood: A qualitative study

Abstract Purpose Definitional perspectives and operational approaches to the concept of resilience vary within the literature; however, little is known of women's opinions on current resilience research, or the philosophical and methodological directions women believe such research should take. This research explored women's perspectives on resilience research in the perinatal period and early motherhood and sought their opinions on the ways in which they believe research should be advanced. Methods Following ethical approval, online interviews were conducted with 14 ethnically and socioeconomically diverse women who were mothers. Findings from a concept analysis on resilience in pregnancy and early motherhood, conducted by the authors, were shared with women before and during the interview. Interviews were organised in sections corresponding to the findings within the concept analysis' four philosophical (Epistemology, Linguistic, Logic, Pragmatic) principles and thematically analysed. Results Epistemology—Women endorsed a dynamic process definition, and viewed resilience as influenced by multilevel, multisystemic processes. Linguistic—Women viewed words such as ‘adaptation’ and ‘adjustment’ as being more active and empowering than the term ‘coping’ in relation to motherhood. Logic—Women were resistant to the predominant operational conceptualisation of resilience as illness absence. Pragmatic—Women were wary of resilience research being used to reduce mental health support for other mothers and families. Conclusions Women provided constructive criticisms on the current state of resilience literature. Women suggested actionable ways in which research may be developed to better align with the epistemological and ethical approaches women want to see in resilience and maternal mental health research. Patient or Public Contribution Women who are mothers and participants in health research were consulted on their views of trends in mental health and resilience research in motherhood.


| INTRODUCTION
There is an ongoing philosophical shift in approaches to mental health research as researchers challenge the prevailing focus on psychopathology in which health is, by default, defined as the absence of an illness or disorder. 1 Included within this shift from pathology to a strengths-based investigation, is the concept of resilience. 2 Although definitions and operationalisation of resilience vary, there is some consensus in the broad understanding of resilience as 'positive adaptation despite adversity'. 3,p.739 Within the maternal mental health literature, the concept of resilience has enjoyed rapid growth in research interest over the past 2 years. 4 Resilience is a logical avenue of mental health investigation considering that women are at greater risk of developing a mental health problem (MHP) in the first postpartum year than they are pre or during pregnancy, 5 and the negative impact of MHPs on child development. 6,7 Additionally, resilience is a point of interest as several studies demonstrate higher levels of anxiety and depression years into motherhood compared to the first postpartum year. [8][9][10] Despite increased interest in resilience in motherhood, women's perspectives on the varying definitions and approaches to measurements are distinctly absent from research. 4 Few studies centre on women's voices and active participation in mental health or resilience research in motherhood, 11 and, in general, little is known of the mental health research that women wish to see conducted. 12 This is concerning as research shows substantial disparities between the health services offered and the supports women wish existed to meet their needs. 13,14 This gap illustrates the need for women to hold an active role in maternal mental health and resilience research, as failure to integrate the community's/research participants' perspectives is likely to result in findings that reflect the priorities, biases and worldviews of the researcher(s) rather than the needs of those whom the research concerns. 15 The current paper focuses on the second of a three-phased research design ( Figure 1). Approaches to resilience differ according to context; however, the extent to which resilience research in motherhood follows or deviates from trends in resilience research in other contexts was unknown. Therefore, the first phase comprised a context-specific concept analysis that described the current state of the literature on resilience in the perinatal period and motherhood (defined as up to 5-years' postpartum). 4 The concept analysis used a principle-based framework 16 which evaluated data according to four philosophical principles: Epistemology, Linguistics, Logic and Pragmatism. The analysis identified that in the context of the perinatal period and early motherhood, (i) research frequently operationalised resilience through illness absence, (ii) there was interchangeable use of associated concepts such as 'coping' and 'adaptation', (iii) measures of positive adaptation were predominately related to the mothering role and (iv) there were few qualitative explorations of women's resilience. 4 The concept analysis was conducted with the intention of informing the current phase of the research. This phase aimed to (i) ascertain women's views on the current methodological and conceptual approaches to resilience in the perinatal period and early motherhood, and (ii) seek their opinions on the ways they believe resilience research should be advanced. Data from the current phase of research is intended to inform (i) researchers of the objectives that women wish to see included in resilience research, and (ii) the development of themes of investigation for in-depth qualitative interviews on resilience in the context of motherhood.

| Participants
The university's Research Ethics Committee granted ethical approval.

Participants in the longitudinal Maternal health And Maternal
Morbidity (MAMMI) study were invited to join online one-to-one interviews. Participation was open to all consenting participants in the study without exclusion criteria. Recruitment took place over two periods; November 2020 and June 2021. Women received a participant information leaflet, a hard-copy consent form for their own records and a link to an online consent form, a summary of the concept analysis findings from phase one and a semistructured F I G U R E 1 Research phases. interview guide. Seven women were interviewed following the first recruitment invitation which used convenience sampling. However, it became apparent that a majority (not all) of respondents to the first invitation were White-Irish, partnered or had a third-level education.
Therefore the researchers issued a second invitation using purposive sampling to recruit women from more ethnically and socioeconomically diverse groups. Participants self-identified if they belonged to one or more of the following groups: ethnic minority (Irish traveller, Black, Asian, ethnic minority groups), migrant (of any ethnicity), seeking/received asylum or refugee status in Ireland, 30 years of age or younger when they had their first child, do not have a graduate (third level) education, experiencing or experienced homelessness in the past 7 years, LGB identifying and single mothers. Seven women were interviewed during the second recruitment period. The participants' children were aged 6 months to 7 years old. Fourteen participants consented to the interview (Table 1).

| Procedure
The research used a qualitative descriptive design. Data were collected using one-to-one semistructured interviews conducted via Microsoft Office Teams or telephone depending on the preference of each participant. The interview guide was developed by the researchers and corresponded to the findings from each of the four (Epistemology, Linguistic, Logic and Pragmatic) principles of the concept analysis ( Table 2). The key findings under each principle were shared with participants, and then they were asked related questions from the interview guide. Participants confirmed that they had shared all they wished to say before the interviewer moved to each consecutive principle. Participants were encouraged to ask questions at any point in the interview. One author (S. H.) undertook all interviews, which averaged 1 h and 18 min (range: 44 min to 1 h 49 min).

| Data analysis and rigour
Interviews were audio-recorded, transcribed, pseudonymised and thematically analysed using Braun and Clarke's methodology 17 and managed using Microsoft Excel. Analysis was conducted concurrently through data collection to encourage an iterative interaction between the data and analysis. 18 To ensure coding consistency and agreement, all researchers independently coded two interviews from each recruitment period, compared and refined codes and themes and discussed confirmatory and negative cases. 18 The data were analysed in sections that corresponded to participants' responses to the findings of each principle (Epistemology, Linguistic, Logic and Pragmatic) of the concept analysis. Findings are presented accordingly.
Each participant received a synopsis of the interview findings, supported by anonymised quotes. They were asked if they felt that the findings were representative of their views and/or if they felt that the researchers had under-or overemphasised findings. This step further integrated participants' involvement within the research process to ensure the credibility of the findings through member checking. 19 Participants received the synopsis via email and a posted hardcopy, and were invited to respond in any way that was most convenient to them (e.g., email or post replies). Four women responded to add to or clarify their views on the research. For example, one woman suggested that the tensions that arise from navigating cultural differences as a migrant woman and mother living in Ireland were an issue of personal importance and suggested it as an avenue for further research. Each participant confirmed that the findings were representative of their views and some mentioned that they felt that the collated findings had captured some issues which they themselves had not given voice to in the interview. Participants' identities have been anonymised for publication and illustrative quotes are presented using pseudonyms.
The researchers are maternal health researchers with backgrounds in psychology, midwifery and mental health nursing. All acknowledge that they are advocates of women's health, participatory/advocacy approaches in research design and conduct and the centrality of women and their voices.  Overall, a majority of women felt that a dynamic perspective most accurately captured the meaning of resilience. The largest portion of the women's conversation related to factors external to the individual. Several described how upbringing, family and culture provided a source and an exemplar of resilience.
T A B L E 2 Lay synopsis of findings from resilience in pregnancy and early motherhood concept analysis and interview guide.
Philosophical principle Lay synopsis of concept analysis findings derived from Hannon et al. 4 Interview guide Epistemological findings: How is resilience defined?
Most often resilience was defined as a trait: trait definitions approach resilience as a set of personal/internal traits which are a stable feature of someone's personality, and these traits help someone to be resilient when they are faced with challenges. In this approach, resilience is often measured using a scale. There is a lot of good research to show how certain aspects of personality are associated with better mental health outcomes during or after adversity. But, it may be difficult to develop resilience-based interventions that can be used with a large number of people if resilience is considered related to individual personalities. Resilience was sometimes defined as a process: process definitions consider resilience to be an ongoing process influenced by multiple individual, contextual, familial, social, environmental, political, economic and cultural factors. These approaches sometimes look at mental health outcomes (low psychopathology, high mental well-being) and/or positive adaptation outcomes (functionality, competence, etc.). The rest of the studies provided an explanation of how they would measure resilience (usually as stable levels of depression or anxiety over time), or did not give a definition of resilience.
What are your thoughts on these definitions and perspectives? Do you agree/disagree with these definitions? How do you define resilience? What perspective should researchers take in resilience research?
Linguistic findings: What kind of language is used in resilience research?
In the perinatal and early motherhood literature, the terms coping or coping strategies, adaptation and adjustment, protection and resistance were commonly used or associated with resilience.
What do you think about these terms? How do you feel they fit into the concept of resilience?
Logical findings: How is resilience in motherhood measured?
(i) Resilience scales: Resilience scales are usually used where researchers take a trait approach to resilience; however, they are rarely used alone and are often used alongside mental health outcome measures. (ii) Mental health outcomes: Depression was the leading mental health outcome of interest in the maternal literature, followed by stress disorders such as PTSD, and anxiety. In most cases, low symptomology or illness absence is considered indicative of resilience as this is an ideal outcome, especially in contexts of adversity. Some studies also included measures for mental well-being, quality of life, self-compassion or psychological flexibility. (iii) Positive adaptation outcomes: Positive outcomes in the perinatal period and early motherhood literature frequently related to a woman's adaptation and competence in the parental role, such as parenting sense of competence or family functioning.
What are your thoughts on the ways that resilience is currently measured? How would you like to see resilience measured in future research? Women considered coping, poised as a question, as restrictive.
They felt there was only one acceptable response, which potentially hindered women's openness about the difficulties of parenthood.
It's not ok as a mother to say that you're not coping, where (it might be seen) that you're not doing a good enough job. (Saoirse) From this viewpoint, coping was seen as a short-term strategy offering temporary distress alleviation; coping was viewed as adequate for survival but inadequate for living or thriving, whereas resilience was aligned with long-term solution finding.
I don't think that coping is the same as resilience, coping is… barely scraping by, coping is surviving….
Coping has to do with staying barely above the water.
Resilience is being able to swim.

| Women's perspectives on the Pragmatic findings
In relation to the application in research, all women endorse efforts to conduct research that centres women's voices as the experts of their lived experiences. They were of the view that inclusion indicated respect and a progressive approach to research. The concept analysis did not reveal any study actively employing resilience or a resilience scale as a method of assessing or screening mental health in the postpartum period or early motherhood, however, it was presented as a suggestion by several authors. Many women questioned the feasibility of mental health assessments at all (regardless of whether the tools used in the assessment are for illhealth or resilience) for two reasons: first, women frequently noted that the interactions they had with HCPs were not amenable to assessment or screening as appointments were short, and focused on their child's health. Second, a number of women described the severe lack of quality, timely service response following disclosure of mental distress.

| DISCUSSION
Few studies centre women's voices and active participation in mental health or resilience research in motherhood, 11 and, there are concerns that this absence may exacerbate gaps between the health services available and the health services that women actually need. 12 Additionally, the absence of public and patient voices as important stakeholders in health research raises concerns about producing research that meets the public's needs. 15 Indeed, the current research illustrates the disparity between the ways in which resilience in motherhood is presently researched and the ways in which women believe resilience should be researched.
Women engaged deeply in appraising empirical research and provided constructive criticisms on the current state of the literature and suggestions for pathways for developing future investigations.
Although women identified a number of personal factors they perceived as associated with resilience, they were concerned that adoption of the trait perspective alone was reductive. This echoes the ethical concerns around trait conceptualisations in the literature. 20 The findings show that women support a biopsychosocial-ecological approach to resilience. 21

| Limitations
The study included participants for whom English was not their first language, however, the study's structure potentially excluded women who did not feel that they possessed sufficient English skills.
Additionally, the research collected minimal socio-demographic information from participants, and while the study included women of differing educational levels and backgrounds, most disclosed having a third-level education qualification. Therefore, the findings should be contemplated in this light, and consider that women who have fewer socioeconomic resources to avail of further education may interpret resilience differently or hold different views as to how resilience should be researched.