“Stop, pause and take a break”: a mixed methods study of the longer-term outcomes of digital emotional wellbeing training for perinatal women Authors: Jacqueline A. Davis, Jeneva L. Ohan, Natasha Bear, Lisa Gibson, Desiree Silva, Susan Prescott, Amy Finlay-Jones Published: 2024 ABSTRACT BackgroundMaternal psychological distress is related to poorer physical and mental health as well as child developmental problems. Interventions that optimise maternal mental health and wellbeing during the “first 1,000 days” of life should have wide-reaching benefits for the mother and her child. Digital mental health interventions (DMHIs) during this critical period have the potential to equip women with enhanced coping strategies that can be applied and maintained in daily life. This study aimed to understand uptake of DMHIs in pregnancy, and the extent participants apply intervention strategies 10–12 months post-participation in a pilot randomised controlled trial (RCT) comparing different digital programs.MethodsA mixed methods triangulation approach was undertaken. We explored demographic data and psychological distress measures in the intervention groups and a comparison group that were offered the intervention but did not participate or did not meet the inclusion criteria based on gestational age (n = 525). Intervention participants (n = 54) were invited to undertake an interview up to 12 months post-intervention to understand ongoing skills application. Fifteen interviews were undertaken using a phenomenological approach to analyse the data.ResultsBaseline psychological distress results showed a significant difference between those who enrolled in the intervention, compared to those who did not, with higher psychological distress (combined), and stress and depression (sub-scale) scores for the intervention groups. Qualitative content analysis identified four main themes relating to long-term engagement: (1) Motivations to use skills; (2) Taking time for oneself; (3) Emotional support; and (4) Improving wellbeing (such as aids sleep and emotional impacts). Within each theme there were several sub-themes, including enablers or barriers to engagement.ConclusionsThis study indicated that women who chose to enrol in DMHIs in pregnancy have greater psychological distress than those who chose not to participate. Many participants in this study continued to apply learnt skills in everyday life, such as breathing exercises. This implies that digital strategies may be effective and applicable longer-term to enable women to apply positive coping skills during critical child developmental periods. Ultimately, this will contribute to designing apps that sustain wellbeing and could be protective in preventing postnatal psychological distress.Trial registrationAustralian New Zealand Clinical Trials Registry Number (ACTRN12620000672954p); (12/06/2020).Supplementary InformationThe online version contains supplementary material available at 10.1186/s12884-024-07002-z. Background The “first 1,000 days” of life, spanning from conception to a child’s second birthday, have been identified as a critical time for development, with implications for lifelong health and wellbeing [1]. Maternal psychological distress – that is, heightened and prolonged feelings of stress, anxiety and depression—is one important influence during this period that has been associated with child developmental problems, including reduced cognitive and language skills, difficult temperament, and increased behavioral and emotional problems during childhood [2–4]. Furthermore, there is growing evidence that maternal psychological distress during pregnancy can lead to changes in fetal physiology and behavior, such as decreased fetal heart rate variability, altered sleep patterns, and increased activity levels [5]. This may be mediated by transplacental neuro-endocrine factors (including cortisol) and immune/inflammatory processes that have the capacity to influence fetal development and physiological responses that persist into adulthood [6, 7]. This direct link between maternal wellbeing and the lifelong health of offspring is the basis of the “developmental origins of health and disease (DOHaD)” hypothesis put forth by David Barker [8, 9], and supported by the much earlier work of Rene Dubos who demonstrated that adversity in the mother’s early environment had long-lived effects on progeny [10]. In addition to the implications for child emotional-behavioural development, maternal psychological distress during pregnancy is also associated with poorer outcomes for the pregnant woman, including more medical visits, obstetric complications, shorter gestation, greater and increased risk for postpartum depression [3, 4], as well as other perinatal mental health issues such as anxiety and postpartum psychosis. This underscores the critical importance of disease prevention through early interventions focused on improving maternal mental health and wellbeing, and supporting mothers in the wider familial and societal contexts, as these would not only mitigate maternal suffering but also minimise the risk of negative health effect transmission to the next generation [11]. This is especially pressing in our current context, as a further consideration is the exacerbation of psychological distress caused by the ongoing COVID-19 pandemic affecting perinatal women and potentially future generations [12, 13]. There are a plethora of evidence-based interventions to ameliorate psychological distress and support wellbeing that are applicable in the perinatal period, ranging from universal [14] to targeted prevention [15]. Digital mental health interventions (DMHIs), delivered online and offline via computer, tablet or smartphone, may be particularly useful during this time given that many pregnant women search the internet for health information [16, 17]. DMHIs have the potential to overcome a range of accessibility barriers to mental health programs [18], such as cost, stigma, time and transport [19]. A systematic review by Ashford et., (2016) [20] indicates that DMHIs targeted at improving mental health can be a promising and advisable form of intervention during the pre and postnatal periods, particularly given issues in accessing timely and flexible care. However, despite promising indications, there are few existing DMHIs tailored to women during and post- pregnancy that specifically target building positive mental health and emotional wellbeing [21]. A recent systematic review by the authors identified a number of DMHIs delivered in the perinatal period to women with sub-clinical mood symptoms that report initial participant engagement, but insufficient reporting of participants’ real-world use of intervention skills [22]. Digital health engagement is complex with many interconnecting factors that affect participants’ enrolment and engagement in the intervention, and application of learnt skills in real world settings [23]. To bridge this knowledge gap in understanding longer-term application of active skills, the barriers and enablers for delivery and sustainability of DMHIs need to be better understood in practice [24]. During 2020–21, our research team undertook a pilot three-arm randomised controlled trial (RCT), Mums Minds Matter (MMM), comparing DMHIs based in contemplative practices: mindfulness training (MT), loving-kindness and compassion training (LKCT) and progressive muscle relaxation (PMR, active control) to improve positive mental health and reduce psychological distress among pregnant women [25]. In their systematic review Agteren et al. [26], emphasised the superior effects of mindfulness-based and multi-component interventions in both clinical and non-clinical populations compared to a range of other psychological wellbeing interventions. Mindfulness-based interventions (MBIs) and compassion-based interventions (CBIs) are intervention approaches based in contemplative practice that target both alleviation of distress and cultivation of wellbeing [27–29], and are associated with positive outcomes for women and families during the perinatal period [30, 31]. Importantly, MBIs and CBIs can be delivered digitally to perinatal women providing easy, cost-effective and universal access for this population group. These interventions are similar in that they rely on regular practice of meditation as an intervention mechanism; however they diverge in the focus of the meditations and related practices. Additionally, there is no real understanding of which components of DMHIs, specifically those based on contemplative practice, women actually use in their daily life and there is a paucity of evidence on their long-term effectiveness [24]. For digital interventions to have long-lasting impact we need to understand what motivates women to undertake the intervention, how women utilise the intervention skills, and what skills and effects are maintained after intervention completion. The CAPE model [32] examines four aspects of intervention engagement: connect, attend, participate and enact. In the perinatal population not all steps are well understood, especially, ‘connect’ and ‘enact’. Much research reports on attrition rates once participants have signed up for an intervention, but we need to characterise motivations for enrolling as the initial step in engagement. Furthermore, a crucial but often overlooked measure of intervention engagement is enacting the skills in daily life; without sustained skills use, there is little opportunity for the intervention to take effect, no matter how high the person’s attendance in the intervention. In this study, we sought to understand pregnant women’s connection (that is motivation to enrol) in contemplative practice based DMHIs, as well as their enactment (that is, their long-term post-intervention strategy use) following birth and during the child’s first year. We also sought to better understand the demographic and clinical characteristics of women who chose to enrol in the intervention. Ethics approval The MMM study was approved by the Ramsay Health Care WA I SA Human Research Ethics Committee (#1927). It received ethics recognition from The University of Western Australia (2020/ET000334). Further ethics approval was provided for follow-up with participants and access to ORIGINS participant data. Methods Setting MMM was nested within a longitudinal, prospective pregnancy cohort study, ORIGINS, that is tracking 10,000 Western Australian families enrolled during pregnancy and followed over the first five years of life. Women were recruited into MMM from the ORIGINS cohort from September 2020 until February 2022. Procedure & study design Since MMM participants are part of ORIGINS, we explored baseline data from an ORIGINS comparison group to examine whether there were baseline demographic and clinical differences between the women who initially engaged in MMM compared to those who did not. Interview data for this study were collected post-intervention, exploring participants’ experiences, ongoing application of skills, and perceived barriers and enablers to practice. Qualitative research methods enable health researchers to delve into questions of meaning, examine social practices and processes, identify barriers and facilitators to change, and discover the reasons for the success or failure of interventions [33]. Comparison Group within ORIGINS All women participating in ORIGINS are given a battery of self-report assessments at multiple timepoints, antenatally and postnatally. These include assessments of the mother’s mental health and wellbeing, such as the Depression Anxiety and Stress Scale (DASS-21) [34], which is completed by participants twice antenatally (at 18 and 36 weeks’ gestation), then postnatally at 6 months, 1 year, then annually every subsequent year. Since all MMM participants consented to be part of ORIGINS, we were able to apply to access additional routinely collected data, specifically data on participants’ psychological distress at 18 weeks’ gestation (or earlier) during the same time period (September 2020 until February 2022). This enabled us to explore any baseline differences in several factors, such as age, parity and socio-economic advantage/disadvantage, as well as psychological distress in the women who chose to participate in MMM, compared to those who abstained or were not eligible for MMM. Demographic data Socio-economic status was determined by Socio-Economic Indexes for Areas (SEIFA)/Index of Relative Socio-Economic Advantage and Disadvantage (IRSAD). Scores were calculated based on the Australia Bureau of Statistics (ABS) data cubes which uses participants’ Postal Area Code (POA) [35]. Psychological distress Psychological distress was measured using the Depression Anxiety and Stress (DASS-21) questionnaire [34]. This measure comprises 21 items that group into three sub-scales of depression, anxiety, and stress. The items are rated over the past week and higher scores indicate higher levels of depression, anxiety and stress. The DASS-21 is widely used and has demonstrated validity in both clinical and community settings. In this study, scores from the DASS-21 were used to calculate psychological distress scores as a total summative score. We also considered the sub-scales for each depression, anxiety and stress. Interview data Participants Eligible women for MMM were pregnant women consented into ORIGINS who attended their first hospital antenatal visit at Joondalup Health Campus or via telehealth (i.e., at 18 + weeks’ gestation). Upon providing consent to participate in MMM, women were stratified by parity and randomised to one of the three digital intervention arms: mindfulness training (MT), loving-kindness and compassion training (LKCT), or progressive muscle relaxation training (PMRT) (active control). Over eight weeks during pregnancy, participants were given access to a weekly psychoeducation module, a daily meditation practice, and a weekly informal or “on-the-spot” practice that mapped to each theme. The full study protocol includes detailed information on the pre-, during and post- battery assessment questionnaires as well as the intervention format and structures [25]. In total, 84 participants were randomised to MMM and an initial study was undertaken by the authors to understand the feasibility of engaging perinatal women in digital contemplative interventions by documenting the experiences of MMM, during and immediately post-intervention [36]. This current study is a longitudinal follow-up with existing MMM participants, 12 months’ post-intervention. We excluded those who undertook the PMRT arm in the original MMM study as we were specifically interested in contemplative practices; that is, mindfulness (MT) and loving kindness and compassion (LKCT) training. Fifty-four participants undertook either LKCT (n = 28) or MT (n = 26). We contacted all 54 participants by email informing them of follow-up interviews on their long-term experience and application of learnt strategies, clarifying that the research team were interested in both positive and negative feedback to help design a future full-scale RCT. We were interested in all participant perspectives, even those who withdrew or did not undertake weekly assessments. A $25(AUD) e-voucher was offered to interview participants. All follow-up contact was undertaken by the lead researcher (JD). We interviewed those who self-selected to participate (n = 15), that is through convenience sampling. Women did not report any reasons why they did not elect to participate but chose not to respond to the invitations. Interviews were semi-structured using an interview topic guide developed by the project team; core questions focused on enactment of learnt skills, perceived benefits or adverse experiences, ongoing wellbeing practices, and general wellbeing needs of women postnatally. A copy of the semi-structured interview questionnaire is provided in Supplementary Files 1. Once participants agreed to be interviewed, an interview time was set at the convenience of the participant and offered via phone or video conference. Prior to interview, the aim of the research was explained, and further verbal consent was provided for the interview to be recorded. The interviewers were not blinded to the intervention arm. The recommended guideline, consolidated criteria for reporting qualitative research (COREQ) [33] was used in this study (see Supplementary File 2). Interviews were conducted by the first author (JD) and the fourth author (LG) and transcribed verbatim, with interim feedback to the project team to review core themes arising from the interviews. Interviews ceased in the absence of new themes, and once authors agreed data had reached saturation point. Data analytical approach In total, there were n = 525 women in the comparison ORIGINS group, n = 28 in the MMM LKCT (Loving Kindness & Compassion Training) group, and n = 26 in the MMM MT (Mindfulness Training) group. All MMM participants in the LKCT and MMM groups were invited to participate in a follow-up interview. Figure 1 indicates the respective participant study groups. A mixed-methods triangulation design was utilised [37] assessing both qualitative and quantitative data to interpret the results. The data were synthesised and triangulated as multiple sources of data generate a more complete understanding [38].Fig. 1Participant study groups Participant study groups Quantitative analysis For the psychological distress scores, we examined differences between the intervention and the comparison groups. Baseline DASS scores were right-skewed, therefore DASS scores are presented as medians and interquartile ranges and differences between groups analysed using a non-parametric (the Kruskal Wallis) test. Categorical data is presented as frequencies and proportions with differences between groups analysed using Fishers exact test. Statistical significance level was set at p < 0.05, and all statistical analysis was performed using Stata v 16 (StataCorp, College Station, Texas). Qualitative analysis The qualitative analysis used a phenomenological approach, examining human experiences through the descriptions provided by the participants. Phenomenology contributes to deeper understanding of lived experiences [33]. The individual interviews meant we had an opportunity to delve into the subtleties of individual experiences, highlighting unique perspectives and nuances. We could then identify common themes among the shared experience. Qualitative data were transcribed, then analysed by the first author by inductively identifying, reviewing and defining higher order themes and sub-themes [39]. A second researcher reviewed the primary analyst’s interpretation of the data, and codes and themes were decided on after discussions between the two reviewers. Exemplar quotes were used to highlight themes. Qualitative data from both groups was analysed together using qualitative content analysis [40] assigning a code to each concept using NVivo (V1.5). Results Comparison group There were no significant differences between the intervention and comparison groups’ demographics, including age, socio-economic status and number of children (see Table 1). Table 1 Comparison and participant groups’ demographic details Comparison vs MMM cohort Comparison Group LKCT Group MT Group P -value (n = 525) (n = 28) (n = 26) Mothers age (at birth) Mean (SD)32.6 (4.6)33.0 (4.0)32.3 (4.6)0.961 Ethnicity Australian Peoples177 (34.1%)9 (36.0%)9 (34.6%)0.422 New Zealand Peoples14 (2.7%)1 (4.0%)0 (0.0%) British178 (34.3%)9 (36.0%)7 (26.9%) Irish27 (5.2%)1 (4.0%)1 (3.8%) European63 (12.1%)2 (8.0%)9 (34.6%) African15 (2.9%)0 (0.0%)0 (0.0%) Asian38 (7.3%)3 (12.0%)0 (0.0%) American (North, South, Central)5 (1.0%)0 (0.0%)0 (0.0%) Other2 (0.4%)0 (0.0%)0 (0.0%) Missingn = 6n = 3n = 0 SES : IRSADa quintiles 1, most disadvantaged/lack of advantage20 (3.8%)0 (0.0%)0 (0.0%)0.490 214 (2.7%)2 (7.1%)2 (7.7%) 397 (18.5%)4 (14.3%)7 (26.9%) 4207 (39.4%)11 (39.3%)8 (30.8%) 5, least disadvantaged/most advantaged187 (35.6%)11 (39.3%)9 (34.6%) Missingn = 0n = 0n = 0 Highest education completed High school124 (23.7%)6 (23.1%)7 (26.9%)0.661 Trade / apprenticeship91 (17.4%)3 (11.5%)1 (3.8%) University264 (50.5%)15 (57.7%)16 (61.5%) Other44 (8.4%)2 (7.7%)2 (7.7%) Missingn = 2n = 2n = 0 Siblings (yes/no) No190 (53.1%)8 (44.4%)7 (41.2%)0.860 Yes168 (46.9%)10 (55.6%)10 (58.8%) Missingn = 167n = 10n = 9 a IRSAD Index of Relative Socio-Economic Advantage and Disadvantage. This index summarises information about the economic and social conditions of people and households within an area, including both relative advantage and disadvantage measures [35] Table 1 Comparison and participant groups’ demographic details Comparison vs MMM cohort Comparison Group LKCT Group MT Group P -value (n = 525) (n = 28) (n = 26) Mothers age (at birth) Mean (SD)32.6 (4.6)33.0 (4.0)32.3 (4.6)0.961 Ethnicity Australian Peoples177 (34.1%)9 (36.0%)9 (34.6%)0.422 New Zealand Peoples14 (2.7%)1 (4.0%)0 (0.0%) British178 (34.3%)9 (36.0%)7 (26.9%) Irish27 (5.2%)1 (4.0%)1 (3.8%) European63 (12.1%)2 (8.0%)9 (34.6%) African15 (2.9%)0 (0.0%)0 (0.0%) Asian38 (7.3%)3 (12.0%)0 (0.0%) American (North, South, Central)5 (1.0%)0 (0.0%)0 (0.0%) Other2 (0.4%)0 (0.0%)0 (0.0%) Missingn = 6n = 3n = 0 SES : IRSADa quintiles 1, most disadvantaged/lack of advantage20 (3.8%)0 (0.0%)0 (0.0%)0.490 214 (2.7%)2 (7.1%)2 (7.7%) 397 (18.5%)4 (14.3%)7 (26.9%) 4207 (39.4%)11 (39.3%)8 (30.8%) 5, least disadvantaged/most advantaged187 (35.6%)11 (39.3%)9 (34.6%) Missingn = 0n = 0n = 0 Highest education completed High school124 (23.7%)6 (23.1%)7 (26.9%)0.661 Trade / apprenticeship91 (17.4%)3 (11.5%)1 (3.8%) University264 (50.5%)15 (57.7%)16 (61.5%) Other44 (8.4%)2 (7.7%)2 (7.7%) Missingn = 2n = 2n = 0 Siblings (yes/no) No190 (53.1%)8 (44.4%)7 (41.2%)0.860 Yes168 (46.9%)10 (55.6%)10 (58.8%) Missingn = 167n = 10n = 9 a IRSAD Index of Relative Socio-Economic Advantage and Disadvantage. This index summarises information about the economic and social conditions of people and households within an area, including both relative advantage and disadvantage measures [35] Comparison and participant groups’ demographic details a IRSAD Index of Relative Socio-Economic Advantage and Disadvantage. This index summarises information about the economic and social conditions of people and households within an area, including both relative advantage and disadvantage measures [35] Psychological distress We compared the baseline psychological distress of participants in the MMM intervention groups, versus those who received no wellbeing intervention (i.e., the comparison ORIGINS group). Table 2 includes DASS-21 scores as subsets for anxiety, depression and stress, and total scores. Table 2Comparison and participant groups’ psychological distress at baseline Comparison vs MMM cohort Comparison Group LKCT Group MT Group P -value (n = 525) (n = 28) (n = 26) DASS anxiety (less than 29 weeks antenatal) Median (aQ1, bQ3)2.0 (0.0, 6.0)3.0 (1.0, 8.0)3.0 (0.0, 8.0)0.346 Missingn = 22n = 4n = 0 DASS depression (less than 29 weeks antenatal) Median (Q1, Q3)2.0 (0.0, 4.0)4.0 (2.0, 8.0)3.0 (2.0, 6.0)0.047 Missingn = 18n = 3n = 0 DASS stress (less than 29 weeks antenatal) Median (Q1, Q3)6.0 (2.0, 10.0)7.0 (3.0, 16.0)10.0 (6.0, 12.0)0.016 Missingn = 25n = 4n = 1 DASS total (less than 29 weeks antenatal) Median (Q1, Q3)12.0 (6.0, 20.0)20.0 (8.0, 34.0)16.0 (10.0, 30.0)0.013 Missingn = 38n = 5n = 1 aInterquartile range 1 = > 25% distribution bInterquartile range 3 = > 75% distribution Table 2Comparison and participant groups’ psychological distress at baseline Comparison vs MMM cohort Comparison Group LKCT Group MT Group P -value (n = 525) (n = 28) (n = 26) DASS anxiety (less than 29 weeks antenatal) Median (aQ1, bQ3)2.0 (0.0, 6.0)3.0 (1.0, 8.0)3.0 (0.0, 8.0)0.346 Missingn = 22n = 4n = 0 DASS depression (less than 29 weeks antenatal) Median (Q1, Q3)2.0 (0.0, 4.0)4.0 (2.0, 8.0)3.0 (2.0, 6.0)0.047 Missingn = 18n = 3n = 0 DASS stress (less than 29 weeks antenatal) Median (Q1, Q3)6.0 (2.0, 10.0)7.0 (3.0, 16.0)10.0 (6.0, 12.0)0.016 Missingn = 25n = 4n = 1 DASS total (less than 29 weeks antenatal) Median (Q1, Q3)12.0 (6.0, 20.0)20.0 (8.0, 34.0)16.0 (10.0, 30.0)0.013 Missingn = 38n = 5n = 1 aInterquartile range 1 = > 25% distribution bInterquartile range 3 = > 75% distribution Comparison and participant groups’ psychological distress at baseline aInterquartile range 1 = > 25% distribution bInterquartile range 3 = > 75% distribution We compared baseline antenatal psychological distress scores using the DASS-21 self-report questionnaire in participants who enrolled in the MMM study compared to those who did not over the same time period. Those in the LKCT intervention group (n = 28) and those in the MT group (n = 26) had significantly higher total summative DASS scores (p = 0.013). Median (IQR): LKCT group = 20 (8–24), MT group = 16 (10–30), ORIGINS’ comparison group = 12 (6–20). Additionally, the stress subset DASS median scores were statistically significant in the MMM groups (p = 0.016) and the depression sub-group (p = 0.047) compared to the comparison group. Interview group: qualitative analysis Analysis of the interviews with participants identified four main themes from their reflections on long-term engagement and enactment of learnt skills: (1) Motivation to use skills; (2) Taking time for oneself; (3) Emotional support; and (4) Improving wellbeing. Within each theme there were sub-themes, some of which were identified by the research team as enablers or barriers to engagement in the training programs. The themes and sub-themes are presented in Fig. 2, along with exemplar quotes. Each theme is described in more detail below and elaborated through direct participant quotes.Fig. 2Themes, subthemes and example quotes Themes, subthemes and example quotes Theme 1: Motivation to use skills This theme describes how participants applied learnt skills in everyday life and used them either in their interactions with their baby or to their own actions. 1a: Connection and interaction with baby Many of the women who participated in either the LKCT or MT training continued to use the skills regularly in their connection and interaction with their babies from pregnancy through to the first year of infancy. “I think there was one meditation that focused on …. like connecting with the baby while you were still pregnant and kind of imagining that relationship and that sort of thing. But I think sometimes when I was feeling a little bit low, in myself, … I felt a little bit disconnected, I suppose, to my daughter. So when I had those feelings, I could reassure myself, I suppose, that I'm not actually disconnected, like it's – you know maybe I'm just distracted and that sort of thing, and we do have a bond in the relationship and that helped to give me some reassurance.” (MMM50-LKCT) “If we go to the park and he really enjoys – I don’t know, he just enjoys sitting there, touching stones and throwing things up in the air and whatever else, I think just not needing to find something to do and just letting him explore.” (MMM24-MT) “I think there was one meditation that focused on …. like connecting with the baby while you were still pregnant and kind of imagining that relationship and that sort of thing. But I think sometimes when I was feeling a little bit low, in myself, … I felt a little bit disconnected, I suppose, to my daughter. So when I had those feelings, I could reassure myself, I suppose, that I'm not actually disconnected, like it's – you know maybe I'm just distracted and that sort of thing, and we do have a bond in the relationship and that helped to give me some reassurance.” (MMM50-LKCT) “If we go to the park and he really enjoys – I don’t know, he just enjoys sitting there, touching stones and throwing things up in the air and whatever else, I think just not needing to find something to do and just letting him explore.” (MMM24-MT) 1b: Increased self-awareness and being present Most mothers mentioned increased self-awareness and being present in the moment. There was a common theme of being present and connection to nature, as indicated in the following participant quotes. “I also really enjoy getting out in nature and being in the moment and kind of being mindful of my environment.” (MMM50- LKCT) “I probably don't do anything for myself but definitely getting outdoors with her is big a thing and just walking.” (MMM43- LKCT) “I make an effort to be, you know, present and – especially if we’re outside to… at least listen to things around us.” (MMM24-MT) “I also really enjoy getting out in nature and being in the moment and kind of being mindful of my environment.” (MMM50- LKCT) “I probably don't do anything for myself but definitely getting outdoors with her is big a thing and just walking.” (MMM43- LKCT) “I make an effort to be, you know, present and – especially if we’re outside to… at least listen to things around us.” (MMM24-MT) 1c: Breathing and relaxation Up to 12 months post-intervention, participants described how the training has continued to enable regular relaxation, predominantly through breathing exercises and visualisaations. “But now, I know sometimes I go to bed and it's just like trying to find that switch off, so I might now put the lights on, you know, do some breathing exercises, go to a happy place, do the visualisations or the techniques, definitely makes the difference.” (MMM43- LKCT) “But now, I know sometimes I go to bed and it's just like trying to find that switch off, so I might now put the lights on, you know, do some breathing exercises, go to a happy place, do the visualisations or the techniques, definitely makes the difference.” (MMM43- LKCT) Theme 2: Taking time for oneself This theme describes time as both an enabler and barrier to practicing skills learnt in the training programs. 2a: Time for oneself Overwhelmingly, participants acknowledged that time for oneself was important, even if it was fleeting, the program reinforced the benefit of fostering personal time. They described this as supporting emotion regulation through having time to focus on themselves and using breathing as a calming strategy to soothe themselves. “I definitely do need a little bit of time to myself to focus on my mental health, because I think that when you have a newborn, you're so focused on them.” (MMM2-LKCT) “I try to spare, at least, 5–10 minutes every day for myself to, you know, reorganise my thoughts and feelings and – and give me some time to – to digest and settle and stuff like that, mentally” (MMM38-LKCT) “I think it’s more allowed me to focus on me a bit more, like it’s okay to go actually, no, I need to step away for 15 minutes. I just need to take a breath and get myself back together.” (MMM32-LKCT) “I definitely do need a little bit of time to myself to focus on my mental health, because I think that when you have a newborn, you're so focused on them.” (MMM2-LKCT) “I try to spare, at least, 5–10 minutes every day for myself to, you know, reorganise my thoughts and feelings and – and give me some time to – to digest and settle and stuff like that, mentally” (MMM38-LKCT) “I think it’s more allowed me to focus on me a bit more, like it’s okay to go actually, no, I need to step away for 15 minutes. I just need to take a breath and get myself back together.” (MMM32-LKCT) 2b: Habit and regular reminders Participants often reflected that the greatest enabler to practice was regularity, in the form of setting aside a routine time, place or setting. Practice therefore becomes a habit, and it was emphasised that establishing this habit before birth helped to sustain it afterwards. “You build that habit before birth and then just continue it afterwards (MMM1-LKCT) “So, I think having a set time in the day, even though it wasn't a specific time, it was just before bed… was helpful.” (MMM50-LKCT) “You build that habit before birth and then just continue it afterwards (MMM1-LKCT) “So, I think having a set time in the day, even though it wasn't a specific time, it was just before bed… was helpful.” (MMM50-LKCT) There were recommendations that the MMM program could be enhanced with more frequent prompts and affirmations, “maybe some text messages or some affirmation you’re doing a good job just to give you that little bit of a boost maybe if you’re having a bad day.” (MMM37-MT). 2c: Too busy New mothers are extremely busy, and many participants simply did not reflect on the program, “So I was just like all consumed with baby stuff for the first eight months.” (MMM54-LKCT). Frequently, participants said that they were just too tired, too busy with the baby or too sleep deprived to continue with the practice. “It’s just I don’t think about it. I’m rushing around, doing things most of the time, working at home, looking after the family.” (MMM3-MT) “It’s just I don’t think about it. I’m rushing around, doing things most of the time, working at home, looking after the family.” (MMM3-MT) While some participants had every intention of continuing with the practice, it was not as high a priority as other activities, “I think I had every intention on going back and looking at them when I had time, but when I have time, all I wanna do is sit down and do nothing, so I never ended up going back and doing them.” (MMM51-MT) Theme 3: Emotional support This theme reflects on the emotional wellbeing postnatal support needs of the participants from the time they completed the antenatal program until the follow-up interview. 31a: Need for postnatal program All participants saw the benefits of providing the training program postnatally, although they differed in their suggested timeframes, some recommending it was immediately available post-birth while others believed they were not ready until after the first three months, or longer. Invariably they saw the need for both an antenatal and postnatal program.“I definitely feel quite passionate about giving women the support early on, you know, like straight after… delivery as well.” (MMM37-MT) “I definitely feel quite passionate about giving women the support early on, you know, like straight after… delivery as well.” (MMM37-MT) 3b: Support network in person or virtual Simply having a support network post birth was fundamentally important to many women. Some women were uncertain where to go for specific support, and conveyed frustration with lack of access to timely help. “It can be one month, two months, three months, whenever or just weeks, days, maybe but there should be… a place that’s open and accessible for everyone and they have to know about it, and they can reach out for help.” (MMM38-LKCT) “It can be one month, two months, three months, whenever or just weeks, days, maybe but there should be… a place that’s open and accessible for everyone and they have to know about it, and they can reach out for help.” (MMM38-LKCT) Participants valued the care of their partners, particularly in supporting their mental health needs, “My husband keeps saying…, “You need to look after yourself to look after her” or have that timeout because I don’t have any time away from her, so just having a break.” (MMM43- LKCT) There were mixed opinions on preferences for in-person or virtual training programs, with most interested in both options. In terms of postnatal services, there was a more consistent leaning towards in person, particularly for breastfeeding advice or baby assessments. “I think, that’s a really bad shift by a lot of health professionals now just thinking that they can monitor mums and babies via telehealth. (MMM33-MT) “I think, that’s a really bad shift by a lot of health professionals now just thinking that they can monitor mums and babies via telehealth. (MMM33-MT) 3c: Isolation and loneliness Several women mentioned their isolation and loneliness in the first few months after giving birth and projected a sense of disconnection. This was more acute for those without local family support. “Talking more or solely about those very early days when you’re completely lost. Mostly when you don’t have like a mum around or another help…, because all our parents are back in Europe, so we were completely on our own and sometimes we felt lost.” (MMM38- LKCT) I found since having [the baby] it to be more isolating than I've ever found before. (MMM43- LKCT) “I felt like after I’d left hospital, I was kind of like left to it. There wasn’t really any follow up support after that. I was just… left.” (MMM3-MT) “Talking more or solely about those very early days when you’re completely lost. Mostly when you don’t have like a mum around or another help…, because all our parents are back in Europe, so we were completely on our own and sometimes we felt lost.” (MMM38- LKCT) I found since having [the baby] it to be more isolating than I've ever found before. (MMM43- LKCT) “I felt like after I’d left hospital, I was kind of like left to it. There wasn’t really any follow up support after that. I was just… left.” (MMM3-MT) 3d: Needs not met by the training program Some participants reflected that support was indeed needed, but that they felt the digital format was inappropriate for them (e.g., needing more intensive intervention, preference not to increase 'screen time' around baby). The training program did not feel relevant, appropriate, or suited to perception of their needs. One participant believed her mental health issues needed “deeper therapy” than the program could offer. Other reasons why participants did not continue with the practices included: forgetting about the program, the format of the program (I, myself, enjoy doing things like games and stuff, rather than… watching a video – I find reading as well much better than a video (MMM51-MT), and a general reluctance to use a digital device, “I don’t wanna be on my phone all the time in front of my daughter.” (MMM54- LKCT) These participants may be identified as requiring further support or support that takes a different format. Theme 4: Improving wellbeing This theme describes other common factors associated with emotional wellbeing during the first year post-birth. 4a: Aids sleep Many participants considered the training played a role in aiding their sleep. One participant discussed her independent discovery of a bedtime sleep stories app that includes general meditations and ‘sleepcasts’ describing in detail a relaxing scene. She used them most nights to fall asleep. 4b: Anxiety and depression A few of the participants discussed their struggles with anxiety and depression, including postnatal depression. Underlying issues were heightened because of COVID, “My anxiety has certainly increased since COVID times, like I think twice about where I’m gonna go and where I’m gonna take my kids. If there’s gonna be big crowds, I’d rather not. (MMM50-LKCT) However, some women actively applied the training tools, such as mindfulness and relaxation techniques, from the program to alleviate their psychological distress, “I had to actively do things that were going to bring my anxiety levels down because like every time I’d get sick or something could happen, automatically my mind goes to the absolute worst-case scenario. So, it was definitely al lot of self-awareness that like I would catch myself going like three anxious thoughts… and just go, “No, I need to go back to mindfulness, I need to go back to relaxation techniques”, or whatever it is and calm myself down.” (MMM2-LKCT) “I had to actively do things that were going to bring my anxiety levels down because like every time I’d get sick or something could happen, automatically my mind goes to the absolute worst-case scenario. So, it was definitely al lot of self-awareness that like I would catch myself going like three anxious thoughts… and just go, “No, I need to go back to mindfulness, I need to go back to relaxation techniques”, or whatever it is and calm myself down.” (MMM2-LKCT) 4c: Impacts of COVID-19 Uniquely, this is a generation of women who experienced pregnancy and the postnatal period during the COVID-19 pandemic. Participants in this study underwent differing lockdown situations, with accessibility restrictions to healthcare, both for them and their child. Many of the women who were interviewed mentioned COVID, situational uncertainties and their heightened anxiety, “I thought COVID was probably the thing that threw a spanner most in the works for me personally.” (MMM17-LK). COVID disrupted the provision of healthcare services, but many women found some level of support through digital platforms including telehealth. However, some participants found this more isolating, “I mean if you look at all those Facebook groups, you would see there’s a lot of women looking for connection and it’s not online… people are looking for in-person.” (MMM33-MT) “I mean if you look at all those Facebook groups, you would see there’s a lot of women looking for connection and it’s not online… people are looking for in-person.” (MMM33-MT) 4d: Unanticipated emotional impact of baby There was a sense of misconception, even frustration, from some women about the unanticipated enormity of pregnancy, birth, and the postnatal period. “One thing that I think women do is a disservice to each other is pretending that everything's fluffy bunnies, roses and rainbows, when pregnancy is f* hard and, postnatally, it's hard too…But I think a lot of people don't talk about the challenges because it's kind of taboo.” (MMM50-LKCT) “I think just having a baby is just so overwhelming generally.” (MMM54- LKCT) “I don’t think people realise how important [emotional wellbeing] really is during pregnancy, and especially that first year after giving birth. It's a massive, massive undertaking, mentally and emotionally. So, anything that I could do to help that is, yeah, more than welcome.” (MMM32-LKCT) “One thing that I think women do is a disservice to each other is pretending that everything's fluffy bunnies, roses and rainbows, when pregnancy is f* hard and, postnatally, it's hard too…But I think a lot of people don't talk about the challenges because it's kind of taboo.” (MMM50-LKCT) “I think just having a baby is just so overwhelming generally.” (MMM54- LKCT) “I don’t think people realise how important [emotional wellbeing] really is during pregnancy, and especially that first year after giving birth. It's a massive, massive undertaking, mentally and emotionally. So, anything that I could do to help that is, yeah, more than welcome.” (MMM32-LKCT) Discussion This study examined initial and sustained engagement of women who participated in a DMHI pilot trial during pregnancy. We examined characteristics of those who elected to enrol in the intervention compared to those who did not, as well as exploring skills they continue to ‘enact’ up to one-year post-intervention. We collected both quantitative data (on psychological distress) and qualitative data (participant interviews about engagement and enactment) to generate insights as to if, and how, participants sustain practice and use skills from DMHIs in the postnatal period. Furthermore, we determined whether there were any notable differences in the type of emotional wellbeing training (mindfulness versus loving kindness and compassion interventions) through our qualitative analysis. Many who participated perceived intervention benefits, and that those who still used the intervention skills were most likely to report taking the time to apply techniques, benefitting themselves and their baby. Participants reported using skills such as breathing exercises, mindfulness, and visualisations to assist with coping and emotional adjustments. This implies that those who could most benefit enrol in programs if they are available. The interviews enabled us to discuss engagement with participants, including with those who did not adhere to the full training programs. For them, other formats may be preferable. Overall, our findings indicated that participants who elected to participate compared to those who abstained (or were not eligible) did not differ on demographics at baseline, but there were differences in their psychological distress levels. Baseline psychological distress results showed a significant difference between those who enrolled in this study, compared to those who did not, with higher psychological distress (combined) and higher stress and depression (sub-scale) scores for the intervention groups. This indicates that participation in the study is skewed towards those who self-select based on perceived need. This is in line with O’Connor et al.’s (2016) systematic review [23] that identify four major themes relating to engagement and recruitment in digital health interventions: personal agency and motivation; personal life and values; engagement and recruitment approach, and intervention quality. Women who participated in MMM were already involved in an existing cohort study (ORIGINS) and therefore may be generally more motivated to participate in studies and/or interventions. The results suggest that participants in both intervention groups continued to apply learnt skills as coping strategies which were identified in the qualitative themes: motivation to use skills, taking time for oneself, and improving wellbeing. This was evidenced through the interview data: the majority of mothers who were interviewed applied the learnt skills either within themselves – such as, making time for themselves, being present, increased self-awareness or breathing exercises—or in their interactions with their child. The interviews did not reveal preferences between the LKCT and MT groups since similar themes evolved from both groups, but there was no within-subject comparison that enabled participants to complete or try all interventions and then state possible preferences for one over the other. However, existing evidence indicates that loving-kindness training may have more significant effects on anxiety and mood relative to mindfulness training [41, 42], specifically in the perinatal period [43]. Self-compassion has been associated with a range of positive psychological outcomes, including lower depression severity, lower trait anxiety and increased positive states [27]. Larger trials over a sustained time period are needed to determine if LKCT can reduce the risk of postnatal depressive symptoms by equipping women with essential coping skills. For future studies, we recommend capturing repeated mental wellbeing measures, during and post-intervention, in a larger intervention cohort to signify long-term wellbeing trajectories. There are more than 10,000 downloadable mental health and wellness apps [44] providing a wide range of tools ranging from information, meditation, symptom tracking, to online coaching, and social support [45]. High attrition rates have hindered research into the prolonged effectiveness of mental health apps [46, 47], and there is still a paucity of evidence on effectiveness of e-health interventions in the perinatal period. Understanding the opinions and experiences of users is critical to designing better mental health apps for more sustained benefits [45]. Our study clearly identifies the key barriers (busy life, lack of priority, not tailored to the individual), and enablers (regular prompts, reminders, affirmations, aids sleep, ongoing accessibility post-birth) to practice over a critical time period in the child’s first year of life. These findings build on our earlier results: that perinatal women want flexible, easily accessible and diverse apps, with daily reminders [48]. Mental health and wellbeing app developers should apply these findings to create engaging apps that generate longer-term behaviour change. A support network is a key factor in protecting against perinatal psychological distress, particularly in facilitating help-seeking behaviour [49]. Our intervention participants perceived benefits of engaging with DMHIs in both the antenatal and postnatal periods, with flexible options to tailor content to their needs, such as for example, the inclusion of sleep meditations. They also express a desire for both digital and in-person support. This is particularly important given that maternal depression may occur before, during pregnancy, and after childbirth thereby making it important to prevent and intervene throughout the entire perinatal period [24]. Women are more vulnerable to mental health problems in the perinatal period, as evidenced by the increased loneliness and isolation reported in our study. However, there are a scarcity of studies that specifically focus on understanding loneliness in parenthood despite the fact that loneliness is associated with increased risks of depression, anxiety and increased stress [50]. A recent systematic review [51] on perinatal mental health literacy found that the ability to recognise mental health problems and to identify relevant symptoms was lacking among both perinatal women and the public. Evidence from our study indicates that pregnant women may recognise their heightened stress and/or depression levels and enrol in DMHIs if available. Thus appropriately targeted, opportunistic interventions should be made available through various sources, including via health professionals [52, 53]. Our study also indicates that perinatal women prefer easily accessible, flexible digital training and may benefit from the inclusion of mental health education and clear signposting for relevant support services (both digital and in-person). For those participants who felt their level of need was not met by the intervention, it might be useful for digital interventions to consider screening and then signpost to more intensive forms of intervention. Mental health problems have been increasing in children and young people intensified by the physical and psychosocial impacts of the COVID-19 pandemic. Fundamentally, we need to understand the earliest possible point at which interventions can effectively prevent the deleterious influences on poor mental health. By equipping families, in particular mothers, with skills to promote wellbeing and the resilience to cope in adversity—especially potentially underestimated challenges of the postpartum period—women will be better positioned to positively influence the next generation through modelling coping ability and positive behaviours. Limitations, strengths and future research This study was undertaken during a time of unprecedented local and global uncertainty. It is not possible to disentangle the impact of the COVID-19 pandemic on this cohort. Over the course of the COVID-19 pandemic, psychological distress in the general population has fluctuated, remaining at elevated levels as of January 2022, compared to 2017 levels [54]. Although not a primary outcome of our study, our results do indicate that women in the perinatal period generally experienced heightened psychological distress when faced with the additional health concerns and future uncertainties posed by the pandemic. This may have contributed to additional recruitment bias (the study may have attracted women experiencing more psychological distress), given that those who consented to participate in the MMM program had elevated DASS-21 scores compared to the non-intervention group. Regardless, there may be a level of participation self-selection bias, as already indicated. A potential limitation is that we did not specifically ask participants why they chose to enrol in this study; it would be beneficial to extract this information in future studies. While embedding MMM within an existing longitudinal cohort study had advantages, the larger ORIGINS population is moderately homogenous with limited diversity in socio-economic status, ethnicity and education. This therefore limits our ability to generalise results, and we recommend a full-scale RCT in a larger and more diverse population. There are limited studies of internet interventions that follow expectant mothers from pregnancy and through childbirth [24]. Additionally, there is a paucity of literature on longer-term enactment and concomitant behaviour change, particularly in perinatal populations. A major strength of the study was our ability to assess participants in a longitudinal pregnancy study, with continued opportunities for follow-up as part of the ORIGINS cohort. An additional strength of the study is in its interview follow-up of participants who withdrew or did not undertake weekly assessments, which allowed for exploration of reasons for attrition in terms of acceptability of intervention. Many studies are unable to reflect on longer-term application of training programs, due to limited accessibility or funding constraints, and high attrition rates [46]. This current study provides new insights on the ‘lived experiences’ of women in the postnatal period, enabling us to determine the types of strategies and interventions that may be most effective and applicable in engaging women to apply contemplative practices in everyday life during critical child developmental periods. Ultimately, this will contribute to enhancing women’s coping skills and sustained wellbeing. Conclusions The notion that “timing is everything” is so much more significant in the developmental context when early life conditions may have lifelong and even trans-generational consequences. Public health policies still require far greater emphasis on the perinatal period, as a critical period for the mother, child and future generations. This includes targeting and timing of integrated interventions, particularly in vulnerable groups [7]. Our study reinforces that many women, particularly those with elevated psychological distress, will be receptive to brief, easily accessible, flexible digital training throughout the perinatal period to enhance their coping skills postnatally. The purpose of these interventions is to halt the progress or risk of psychological distress, and thus identifying those already on the trajectory is also important [18]. In summary, providing women access to effective, enduring coping skills is fundamental in halting the escalating poor mental health pandemic in contemporary societies. Supplementary Information Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2.