Findings

Process evaluation - Evaluation of parenting interventions for those with additional health and social care needs during pregnancy: THRIVE a multi-arm RCT with embedded economic and process components - NCBI Bookshelf An official website of the United States government Here's how you know The .gov means it's official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site. The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely. Log in Show account info Close Account Logged in as: username Dashboard Publications Account settings Log out Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation Bookshelf Search database Books All Databases Assembly Biocollections BioProject BioSample Books ClinVar Conserved Domains dbVar Gene Genome GEO DataSets GEO Profiles GTR Identical Protein Groups MedGen MeSH NLM Catalog Nucleotide OMIM PMC Protein Protein Clusters Protein Family Models PubChem BioAssay PubChem Compound PubChem Substance PubMed SNP SRA Structure Taxonomy ToolKit ToolKitAll ToolKitBookgh Search term Search Browse Titles Advanced Help Disclaimer NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health. Henderson M, Wittkowski A, Buston K, et al. Evaluation of parenting interventions for those with additional health and social care needs during pregnancy: THRIVE a multi-arm RCT with embedded economic and process components. Southampton (UK): National Institute for Health and Care Research; 2025 May. (Public Health Research, No. 13.04.) Evaluation of parenting interventions for those with additional health and social care needs during pregnancy: THRIVE a multi-arm RCT with embedded economic and process components. Show details Public Health Research, No. 13.04. Henderson M, Wittkowski A, Buston K, et al. Southampton (UK): National Institute for Health and Care Research ; 2025 May. Contents Search term < Prev Next > Chapter 6 Process evaluation Introduction This chapter reports on key findings relating to three of the main research questions we sought to examine: What factors affected recruitment to the trial? How faithfully were MB and ETPB implemented in relation to developer manual/protocols and training? What were the mechanisms by which ETPB and MB worked, if they did; who did they work for, how and in what contexts? We consider the perceived benefits, if any, that ETPB and MB add to CAU , and describe the contextual factors that may have influenced their functioning. We also examine the extent to which CAU alone appeared to meet participants’ needs. For more detail on the interventions’ theory of change see Chapter 2 . For more information regarding the realist evaluation methodology and numbers interviewed, see Chapter 3 , process evaluation , the published process evaluation protocol 63 and see Appendix 7 . Findings Each of the following sections summarise findings for key questions we posed about recruitment, implementation fidelity, and how MB and ETPB ’s components worked, or not, for whom and in what contexts. Key question 1: what factors affected recruitment to THRIVE? Overview The PE examined how the practitioner-led model of recruitment worked, and how different organisational and managerial contexts influenced referrals. A paper presenting a more detailed analysis of these recruitment data has been published. 73 Differing perspectives on what influenced recruitment Referring practitioners Referring practitioners (RPs) thought that slowness in midwife (MW)-led recruitment was inevitable. Any one clinic might include zero to a handful of women eligible for THRIVE and among these it was not always possible, or sensitive, to broach THRIVE . The length of booking appointments varied greatly across the different hospitals studied in A&A and GGC , ranging from 30 to 90 minutes, meaning some RPs had significantly less time than others to raise THRIVE with potential participants. One of the RPs (from a clinic with 90-minute appointments) observed that her colleagues simply struggled to ‘keep THRIVE in mind’, which led to some patients ‘slipping through the net’. It was suggested that the THRIVE team use visual methods around clinics to prompt RPs to refer. This RP also had taken on a self-created ‘champion’ role, reminding colleagues to keep on referring and double-checked patient records to ensure that no eligible women were missed. Additional needs were also not always apparent to MWs at the booking stage. Some RPs said they followed their ‘instincts’ early on, picking up on visual cues of ‘vulnerability’, although others were cautious about making too many assumptions based on appearances. One of the RPs observed that mental ill health sometimes remained ‘hidden’ among more affluent patients. Midwifery senior management teams Referring practitioners were perceived by midwifery senior management teams (MSMTs) to have ‘got into the mindset’ of targeting the most marginalised pregnant women from an early stage in recruitment and found it difficult to adjust when the recruitment strategy altered mid-way through the trial (the THRIVE team sought to broaden recruitment to include women on red, highest level of needs, and amber, women with needs, but less severe than those on the red, pathways). This appeared to be compounded by a change in A&A ’s services from a ‘Vulnerable in Pregnancy’ service to ‘Safeguarding in Pregnancy’, which, under new leadership, developed a more overt child protection remit. This was felt to heavily influence the kind of referrals made from the region. RPs were spread across five community hubs in A&A , which, from the perspective of the MSMT , sometimes made it difficult to oversee and co-ordinate recruitment efforts locally. The introduction of a new electronic patient record system was perceived to negatively impact recruitment at GGC , at least while practitioners transitioned. MWs had to spend more time updating records on two systems during consultations, leaving less time to cover the routine checks/pursue discussions. Research nurses Research nurses (RNs) identified that MWs were not always able to approach eligible mothers-to-be and speculated that this could be due to lack of time. RNs identified reception staff, healthcare assistants and nursing Auxiliaries as having potentially supportive roles for future recruitment, although willingness to help was perceived to vary across clinics. RNs’ experiences of informally ‘screening’ patients in the waiting room reinforced ideas, articulated by RPs, that there might be a visual element to identifying someone as a ‘ THRIVE person’. Vulnerabilities were perceived by some RNs to be embodied, making some mothers-to-be visibly identifiable as having additional needs. How did trial organisation impact recruitment? Trial recruitment began in February 2014. As described previously (see Chapter 2 , Enhanced Triple P for Baby ), recruitment was frozen in NHS A&A for 10 months due to delays caused by revisions to the ETPB programme. A total of 83 women were recruited from NHS GGC during this time and groups continued to be delivered. MSMTs from both regions told us that many of their colleagues interpreted the perceived ‘halt in recruitment’ as meaning recruitment to the trial had ceased entirely. This idea seemed difficult to dislodge even though the THRIVE team worked hard to re-engage clinics. The introduction of THRIVE RNs in clinics was viewed to be a helpful reminder of the continuing need to refer. RPs suggested that motivation, and a sense of personal responsibility for referral to the trial, might be improved if the THRIVE team offered regular updates about the trial status and relayed how their clinic’s/RPs individual contribution had helped with the overall recruitment effort. Participants Who was readily recruited, and why? Most participants viewed the initial invitation to participate in THRIVE as an implicit suggestion that they needed ‘educating’ about parenting. First-time mothers-to-be appeared to be more readily recruited on this basis, perceiving participation as an opportunity to build practical knowledge and skills. Those who thought they had lacked positive maternal role models in childhood appeared particularly receptive to an offer of extra support with parenting. There was also optimism about the potential benefits of group participation among some who had poor family support, due to estrangement or migration. They were hopeful that allocation to an intervention arm might help them establish a link to a supportive network of mothers in their area. Some who were currently in conflict with relatives who had taken on caring responsibilities for their children also viewed involvement in the trial positively. Involvement in THRIVE was perceived as a way of gaining impartial feedback about their parenting, if allocated to an intervention arm, developing new skills and rebuilding confidence in their abilities to parent. Who was reluctant, and why? Several mothers who already had children said they had been resistant to the idea that they might need to acquire new knowledge and skills about parenting. This ‘resistant’ group included some who had had previous children removed due to maltreatment, but they nevertheless decided to participate. Other barriers to recruitment were anxiety about joining groups and the absence of childcare provision. Most participants lacked supportive social networks and the material resources to pay for help. Partner participation in ETPB may have been higher if childcare had been provided (several partners provided child care while mothers-to-be with other children attended groups). Did child protection concerns/possible removal of the baby influence participation? Time point 1 (T1) and time point 2 (T2) interviews with participants identified an inadvertent coercive element to trial participation connected to social work (SW) involvement. Participants who had previously had children removed were fearful of being viewed by SW as ‘unco-operative’ if they declined the offer of participating in a trial offering the potential for parenting support. A few participants therefore said they felt they had to agree to participate in THRIVE when invited, for fear that declining might be ‘used (by SW) against me’: Straight away I said aye, because I don’t want my wean took away fae me. They cannae use that against me: that I ’ve no’ done parenting classes when I have. M16 ETPB A&A The analysis for our published recruitment paper 73 concluded it was important that RNs and research staff identified and worked ‘closely with key groups of referrers. This enabled recruitment activities to adapt to changes in maternity settings/policies throughout the study period. And identifying the most appropriate time and setting at which to discuss recruitment to the study’. We noted that there was a strong association between the time available for THRIVE researchers to be involved in recruitment and the volume of monthly referrals into THRIVE . Key question 2: how faithfully were Mellow Bumps and Enhanced Triple P for Baby implemented? In this section we address each of the key questions we posed about implementation fidelity of ETPB and MB and the contextual factors affecting delivery. A detailed report on fidelity, along with supporting data, is to be submitted for publication elsewhere. 11 What programme content was consistently covered across groups? The ETPB elements that were delivered faithfully included (intended outcomes are italicised): survival skills, positive parenting techniques, flexible routines, and monitoring baby’s behaviour, intended to develop new parenting skills ; creating a safe and nurturing environment, to offer new knowledge about infant development ; and parenting traps, coping statements and abdominal breathing, to improve self-awareness of feelings and behaviours when interacting with infants . The MB elements that were delivered faithfully included: a warm welcome to the group, provision of refreshments, and a non-judgemental atmosphere, intended to improve nurturance and reflective functioning; delivery of ‘my island’ and planning future support, theorised to improve current support for parenting ; ‘Mum in a Million’ quiz to improve skills for interactions with infants ; ‘Brave New World’ and Imitation DVDs, theorised to increase knowledge about infant development ; ‘circles of support’ exercise and ‘ghosts from the past’, theorised to improve maternal representation of early attachment relationships . How did delivery vary across groups for particular elements of Mellow Bumps/Enhanced Triple P for Baby? We noted variation for both interventions in the delivery of pre-group meet sessions, partner support, and postnatal support. For ETPB , time allocated to discussion about additional needs varied, often exceeding the suggested timings, to address the complex relationship between parenting and additional needs; for example, having a low income and/or housing problems. For MB , there was variation in the delivery of some of the more nuanced elements of the programme. Elements important to attachment parenting, such as baby cue cards and video clips, designed to teach about responsive parenting and reading baby cues, were sometimes missed out to save time. A few of the ETPB facilitators added an additional facilitator-led planning element to arrange further health and social care support for those perceived to have the most intensive needs. This appeared to help facilitators manage their own anxieties about particular participants, but undermined ETPB ’s underpinning theory of self-regulation. Demonstrations of how to change nappies and bath a baby were also included in some ETPB groups to address the needs of participants with additional health and social care needs, very late in pregnancy, who had missed CAU appointments ordinarily covering this. In MB a midwife was sometimes invited to answer questions about labour, which was not an intended part of the intervention. What content, if any, was missed out, or not delivered as intended, and why? Mellow Bump’s guided relaxation activity was not delivered as theorised due to lack of confidence about delivering this element, and personal beliefs about, and experiences of, relaxation, which led to the activity being undervalued by some (this was delivered better during the course of the MB intervention, as this issue was identified, and extra support provided through practitioner supervision). The community-based activity was perceived by a MB representative as being poorly implemented. This was because facilitators typically delivered this as a nurturing activity, for example inviting a therapist to deliver massages or other treatments, rather than delivering a low-cost activity intended to increase participants’ confidence about engaging in community-based activities that improve parenting support. However, the MB manual suggested facilitators could organise a pamper session as an alternative option for week 5, so facilitators were, strictly speaking, delivering this element as directed. Macro factors influencing implementation fidelity How did the characteristics of venues affect delivery? Sites for intervention delivery included: teaching rooms in maternity hospitals; community meeting rooms in supermarkets; health centres; church halls; civic and community centres; and a nursery school. Characteristics of some of these venues made intervention delivery more challenging, such as noise, accessibility by non-participants, or lack of comfort. These issues were particularly problematic for MB , given facilitators were trying to create a relaxing, nurturing environment and a closed, safe space. Alternative room spaces were arranged if any were felt to be unsuitable. MB ’s reflective consultant felt that ‘psychological (rather than physical) comfort’ was probably the more important element for delivery, so seemed satisfied that venues were appropriate. How did transport affect implementation? THRIVE was funded to cover the costs of transporting participants to and from group venues by taxis. This approach provided equal access to travel to group sessions, regardless of material resources, and was in keeping with the philosophy of MB , in particular, to nurture participants and reduce maternal stress. This appeared to work as intended for most. However, for a few participants who had experienced sexual violence, travelling alone with a strange taxi driver (usually male) unintentionally caused stress (disclosed informally during participant observation). Staggered late arrivals (due to roadworks, late pick-ups, etc.) presented particular challenges for facilitators. Repetition of material was not included in MB timings, so this sometimes impacted on delivery of other content. However, disruption would likely have been far greater had this transport provision not been made. Social class and status Social inequalities were observed to be imbued within, and constructed through, group interactions. There were a few occasions when participants’ assessments of their own, and others’, moral worthiness counteracted facilitators’ best efforts to create a non-judgemental atmosphere. This included the kind of additional needs that were thought worthy of sympathy or support and who might be considered a ‘good’ or ‘bad’ mother. Achieving a supportive atmosphere appeared to depend on the whole group co-operating and helping to co-construct it, rather than this solely resting on the ‘success’ or ‘failure’ of the facilitator to deliver this element correctly. Micro factors influencing implementation fidelity Facilitators’ delivery skills One of the main contrasts between the two interventions is that MB is intended to have a ‘flat’ structure, with no one being considered an expert or an ‘instructor’, whereas Triple P tends to be more didactic, with a clearer facilitator/participant divide. As the group dynamic is also considered important to the theorised functioning of MB , this intervention is also much more dependent on the skills of the facilitator to encourage participants to form and function as a group. The emphasis of ETPB appeared to be primarily on the programme content. Most MB facilitators lacked experience in group delivery at the start. It understandably took time for them to become fluent in the programme and gain the skills needed to get groups functioning as intended. However, we observed good progress with this over time. For facilitators without a therapeutic background (the majority in this study), delivery might have been improved with clarification about, and an opportunity to practice, how to achieve a nurturing, safe, closed environment (this was addressed through supervision). In ETPB there was evidence of reduced implementation quality when facilitators delivered the same content to multiple groups in quick succession: a marked decline in enthusiasm was noted when a pair of ETPB facilitators were observed at two different time points. Other facilitators reported difficulties with transitioning from managing the group dynamic to delivering one-to-one support for individuals with complex problems. In peer supervision sessions it was suggested that support with maintaining professional boundaries might be helpful for facilitators delivering the intervention to those with additional needs. How consistent are programme materials/protocols in directing facilitators what to deliver? The ETPB facilitator manual and implementation pack, including presentation slides, DVD clips, and a script, provided unambiguous direction on how to deliver the core components in antenatal groups. However, there were inconsistencies across different written materials in the way the number of postnatal sessions, timings, and mode of delivery were described in 2014. Refresher training, delivered in 2015, clarified some programme changes, but the written materials were not revised. Direction about how to implement the postnatal element was vague regarding the mode of delivery and how facilitators were supposed to identify what to focus the home-based discussions on. The MB facilitator pack provided all the DVD clips and materials needed to support weekly sessions. The facilitators’ manual provided a skeleton outline of the programme, with suggested timings. MB developers wished to avoid being prescriptive about delivery, explaining to facilitators that the intervention was ‘more than the manual’. Rather, key components ought to be brought to life through the interaction between participants and facilitators and overall group functioning. However, this did mean that it took time for facilitators to gain confidence in fluent delivery and to develop the narrative to accompany activities, for which there was little direction. Facilitators’ understanding of elements and theory Enhanced Triple P for Baby facilitators were guided through the key elements of the antenatal sessions by a structured set of materials, that is presentation slides and a facilitator’s script, disguising whether there was a lack of understanding at an individual level (it appeared not). There appeared to be a good understanding of the purpose of ETPB home-based visits, but facilitators lacked clarity about how to execute these sessions, which resulted in variation in delivery. When uncertain of how to proceed, facilitators sometimes slipped into more familiar professional roles, for example, conducting the kind of conversation they would have as a health visitor. Mellow Bumps observations revealed that some facilitators lacked understanding of key elements of the programme, including the potential benefits of stress-reduction exercises. There appeared to be some confusion about the difference between relaxing activities, such as watching television, and the specific benefits of stress-reduction exercises. What directions were facilitators given about the adaptability of the programme? Triple P made late modifications to ETPB in response to feasibility trial findings led by the University of Queensland. Their readiness to make further necessary modifications to the programme, based on facilitator feedback, was communicated regularly in training. Modifications to the recommended mode of delivery for postnatal home-based visits were allowed, providing this encouraged participation by reluctant mothers, for example, offering telephone support if home-based visits were not desired. Mellow Bumps had some explicit flexibility built into the programme to enable facilitators to adapt delivery if appropriate, for example, dropping ‘Ghosts from the Past’ if participants found the activity too stressful. However, beyond the explicit components with flexibility, MB facilitators were directed to discussing any further adaptations with a reflective consultant prior to implementation. What role did peer-assisted support (Enhanced Triple P for Baby) and reflective sessions with a consultant (Mellow Bumps) have in implementation fidelity? Enhanced Triple P for Baby’s consultant suggested that faithful implementation was more likely when facilitators were embedded within a supportive network that gave them access to informal support, for example, discussion about group delivery with a colleague over lunch. However, none of the facilitators who delivered ETPB worked together or lived within close proximity to each other to benefit in this way. Enhanced Triple P for Baby facilitators were largely reliant on their pairings for support, as most appeared to struggle to arrange formal peer support outside of group delivery time. A member of the THRIVE team organised one peer support group, attended by five ETPB facilitators in 2015, although support was never taken up again. Mellow Bumps supervision was perceived, by the researcher, consultant and facilitators, to have been critical for correcting poor implementation of relaxation activities and for addressing some conflict in facilitator pairings that impaired implementation. However, supervisory input came relatively late. Mellow Parenting and the THRIVE team reminded facilitators regularly to take up supervision, but facilitators had so many commitments beyond THRIVE they often struggled to find the time, outside of groups, to set these sessions up. Lack of uptake might suggest MB facilitators recruited to THRIVE valued supervisory sessions less than practitioners with a therapeutic background would, as the latter would be used to undergoing clinical supervision. Some of the MB facilitators suggested that embedding reflective sessions at the end of group sessions, particularly early on in implementation, might improve uptake. To what extent did facilitator pairings affect delivery? Both MB and ETPB expected that two facilitators would deliver each group. Among ETPB facilitators, less-experienced facilitators paired themselves with those experienced in delivering Triple P variants, which helped build confidence. However, observation of ETPB training indicated that tensions began to emerge among facilitators as early as day 2; some of the facilitators began to express strong views on who they would/would not like to be paired with. Most of the MB facilitators were new to the MB programme and appeared to be developing their skills at a similar rate. Although pairings were reported to be supportive in the main, a few tensions were observed independently by the researcher and the MB reflective consultant, for example, disagreements over whether or not to deliver relaxation activities. The MB consultant observed that a change in one of the MB pairings significantly improved implementation of relaxation activities in future groups, a view supported by one of the facilitators in question. How did facilitators manage Mellow Bumps/Enhanced Triple P for Baby with other commitments? Other commitments, beyond delivery of ETPB and MB , had an impact on how much facilitators could do outside of the two hours they scheduled to deliver groups, despite being told that extra time could be costed and claimed for. ETPB facilitators had reflected a lot, during training, on the work they would have to undertake to become proficient in the programme. They clearly invested a lot of time in preparation, evident from their annotated manuals and familiarity with slides. With the MB training, less time was devoted to the work MB facilitators would have to do in advance to be ready to deliver groups. MB facilitators were observed, in two of the three groups selected for participant observation, to be disorganised at times and visibly stressed about group delivery. Discussion/disagreements about who would deliver what sometimes took place in front of participants. Participants interviewed about their experiences of these early groups joked about their chaotic nature and the technical issues that seemed to plague delivery in early groups. One participant was highly critical of what happened, describing her experience of the group as the antithesis of a nurturing, relaxing environment. This was observed to have improved significantly by the time the last observed group was delivered, but it highlights the importance of building practice time into training and suggesting ways of preparing for delivery. What role did the THRIVE research team have in implementation? The THRIVE research team took on a greater role in supporting implementation than had been anticipated. The extent of their involvement included: arranging venues, liaising with participants and facilitators to set up groups, arranging transport for participants, and issuing materials and equipment to facilitators. At first the THRIVE team also passed on regular reminders to MB facilitators, sent by MB ’s principal developer, to take up supervision. Subsequently, THRIVE ’s Independent Data Monitoring Committee directed the team not to intervene if it became apparent that elements were not being delivered faithfully. The analysis of implementation fidelity shows how faithfully, or not, each element of ETPB and MB components were implemented in the context of this trial. The antenatal components of ETPB appeared to be easier to deliver felicitously than MB ’s equivalent sessions, although delivery of ETPB postnatal sessions varied greatly because of inconsistencies in written/verbal directions. Enhanced Triple P for Baby’s highly structured antenatal programme, with presentation slides, linked DVD clips, and a script, made it possible for content to be covered consistently in most groups, although contextual factors sometimes impacted on timings, resulting in minor modifications. Key question 3: what are the mechanisms by which the interventions work, if they do, who do they work for, and how? In this section we present findings on how ETPB and MB work, or not, for whom, and in what contexts. This is based on early analyses, other than the findings on how the reflective component of MB functions, which Buston et al . have analysed more thoroughly. 119 How well do participants like the interventions? Among the 112 (27.7%) out of the 404 participants randomised to ETPB or MB (participants who completed post intervention evaluation forms), 100% said they ‘strongly agreed’ or ‘agreed’ that the programme was enjoyable, though it is likely that participation was biased towards those more positive about the interventions. Further analysis of the qualitative data is planned, which will help better understand the elements of ETPB and MB participants found most/least engaging, and why. How do participants respond to knowledge on infant and child development in Enhanced Triple P for Baby and Mellow Bumps? Early analysis of T2 interviews suggests that participants regarded one of the strengths of ETPB to be developing parenting skills and strategies, underpinned by a knowledge of child development. Such knowledge was, generally, regarded as helpful, though it was the strategies and skills which were the highlight for most (see below). With MB , its aims were seen as less concrete and providing information on infant and child development was less often referred to as a strength. How do participants respond to parenting skills, partner skills content and home-based practice in Enhanced Triple P for Baby? Parenting skills and partner skills content were a highlight among the ETPB participants we interviewed. Most identified parenting skills development as a key, and helpful, feature of ETPB , being able to talk about strategies learnt, and often already used in helpful ways by the time of the T2 interviews, for example to monitor baby sleep and crying. Home visits were criticised for being too infrequent. Some women felt that they came too late and would have been more helpful when their baby was younger. Participants’ perceptions of partners’ involvement in the programmes Although not all the women whose partner attended groups reported that he had found it useful, the women generally appreciated this aspect of ETPB . For some it had been very valuable in setting the agenda for them to talk and agree on aspects of caring for the baby. However, a few participants, who described an unequal balance of power in their relationships, highlighted that there could be negative consequences for them personally if they attempted to renegotiate the household division of labour. Post-intervention evaluation forms indicated that 56% of ETPB partners attended the session(s) open to them compared to 8% of MB partners. The optional partner session for MB was only held in 3 of the 28 groups, this was the predominantly the participants’ choice. However, some partners were observed to accompany partners to sessions, sitting outside to offer emotional support at break times. A few were allowed to be present in the room when participants expressed particular anxiety about joining the group. Some MB participants felt that having partner involvement in the sorts of things covered by the group would have been very useful. However, participants were highly sensitive to the needs of others in their group who might not have had a partner or who had experienced problems, such as domestic abuse. This appeared to impact on their decision-making about whether or not to include partners in the session open to them. How do participants respond to the nurturing aspects of Mellow Bumps, along with self-care, relaxation, and planned social activities in Mellow Bumps? Time point 1 interviews suggest that the informal nurturing included in MB , such as the food and drink offered, were very important to those attending. Party foods, for example, cakes, biscuits, crisps, had particular appeal, although provision did not fit with the healthy eating and self-care ethos of MB . Participants were given small gifts of a bubble bath, body lotion, and a £5 gift voucher to buy a pampering treat, to encourage small changes in self-care practices. The earliest T1 interviews indicated that suggested self-care activities, like taking a soak in the bath, or massaging their bumps at night, were being taken up by participants at home. However, observation of later selected groups suggested that gifts were sometimes handed out with no discussion about how the items might be used to improve self-care/encourage relaxation, so uptake may have been variable. Few participants commented on their experiences of, or views on, the more formal aspects of relaxation that were also supposed to be a feature of MB , reflecting poor implementation of the stress-reduction exercise in the majority of groups. How do participants respond to exploration of their past and present difficulties in Mellow Bumps? A paper focusing on the reflective component of MB has been published 119 which focuses largely on the implementation of the ‘My Island’ and ‘Ghosts from the Past’ exercises. It explores the women’s engagement with these and identifies mechanisms of change. The analysis suggested that the process of reflection had the same benefits whether focused on the past or on present difficulties. 119 Participants tended to like the reflective work and engaged with it to varying degrees, dependent largely on the coherence of the group. Attendance for sessions 3 and 4, when these exercises took place, was lower than attendance for other sessions though it is not thought to be because of anticipation of this work. Patchy attendance compromised the coherence of some groups and the development of rapport, which is key to delivering reflective exercises. This made it harder for the women to feel comfortable engaging in reflection. Where there was a coherent group, there was evidence of relief through unburdening, empowerment through support given and received, reduced social isolation through sharing anxieties and increased control through self-care advice. A minority of participants with the greatest health and social care needs experienced emotional barriers to engaging with the reflective work. How does the experience of group sessions contribute to, or inhibit, the change mechanisms? The analysis of the reflective component of MB touches on how the experience of group sessions was important to whether stress was reduced or not. 119 With MB it was difficult to pinpoint precisely what made some groups gel, when others did not. Group heterogeneity was related to the trial design – block randomisation stratified by severity of additional needs and substance use – and prompted variable responses. While some of the women appreciated being placed in groups with women that offered contrasting experiences, in terms of age, parity, and additional needs, others – most notably those who felt they might be stigmatised if their histories of substance use/removal of children became known – felt silenced and somewhat alienated when they were placed in groups with others perceived to have ‘normal’ pregnancy pathways. Observations of selected groups also identified some participants with learning disabilities being constructed as ‘other’ to the rest of the group, when they expressed thoughts or exhibited behaviour that was perceived to be atypical. Analysis of implementation fidelity suggests that group facilitation skills, which took time for MB facilitators to develop, may have shaped, particularly in early sessions, how the group functioned and the overall atmosphere. With ETPB , T2 interviews as well as post-session evaluation forms also indicate that group context was important. If participants gelled, this was highly valued by those interviewed. The group was also important in ETPB in supporting learning strategies to use with baby/the couple relationship. Among the participants who completed post-intervention evaluation forms, the practitioners were rated as being ‘extremely’ friendly by 92% (57/62) of those attending ETPB and 90% (46/51) of those attending MB . The other participants in the group were rated as being ‘extremely’ friendly by 59% (37/63) of ETPB participants and 50% (30/51) MB participants. A total of 63% (37/59) of ETPB participants and 74% (35/47) of MB participants reported becoming friends with at least one other woman in the group, though it is not known how many kept in touch following completion of the group sessions: qualitative interviews suggested very few. Participants from one of the earliest MB groups did report forging close bonds, cultivated outside of the group through social media and regular face-to-face meet-ups. By T2 these relationships had broken down and had, in fact, became rather acrimonious. This resulted in some of the MB facilitators cautioning participants about the use of social media to maintain contact outside of groups. By contrast, another group of participants who had attended ETPB were able to maintain the friendships they first formed at the group. They met regularly and used their ongoing encounters to ‘troubleshoot’ anxieties. Preliminary analyses suggest that group interactions may have affected outcomes more with MB than with ETPB . How do participants engage with the programmes? Generally, data suggest fairly high levels of engagement, though this took different forms. Some of the facilitators worried that non-attendance at subsequent sessions suggested a lack of engagement with the programme, or a lack of rapport with facilitators and/or group members. However, interviews with non-attenders and those who had to miss particular sessions indicate that reasons for absence were complex and usually had little to do with the programmes. Interviews with a small number of participants who did not attend any group sessions, and others who had missed one or more antenatal group sessions, revealed that non-attendance was due to issues, such as sickness during pregnancy, not being able to get time off work, having to care for a sick child, and, most commonly, feeling too anxious to attend. What benefits did social interaction within the group offer participants? Are there any unintended consequences? The data suggest that the group delivery context was one of the most valued aspects of taking part for the participants. Specifically, in relation to understanding how the reflective work may have been effective in reducing participants’ stress, many participants reported enjoying the social aspect of getting together with other participants, and the facilitators, on a weekly basis. 119 Some appeared to have few other sources of social support in their lives, though even for those who did the group was sometimes seen as beneficial because the other participants were at the same stage of pregnancy as them. Analyses uncovered a small number of cases where participants, with the greatest additional needs, were marginalised by the MB group, which may potentially have caused harm and increased stress (see How does composition of the group in terms of different additional needs affect group dynamics? ). 119 The social benefits of MB ’s antenatal and postnatal group sessions, as perceived by participants with a range of different circumstances, included: the rare opportunity to have fun an all-female space (enjoyed by many, although some valued the inclusion of fathers) good to share excitement/compare notes with others at the same stage of pregnancy sharing problems was perceived, by some, to be therapeutic attendance at a group offered breathing space; a break from everyday routines and responsibilities getting out of the house reduced loneliness and was perceived to improve mood it offered ‘time out’ for oneself it built confidence (for some) about group attendance. Enhanced Triple P for Baby participants reported distinct benefits they felt were derived from the social aspect of antenatal and postnatal groups. Enhanced Triple P for Baby groups were perceived, by most who attended them, to provide a productive social learning environment. Social interactions between participants and facilitators and between themselves were used to discuss, plan, and consider how they might implement practical techniques to help with parenting. Examples of discussions perceived to be most productive included: challenging notions of the ‘perfect’ mother/parent; how to settle the baby and remain calm; use of behavioural charts to help manage sleep and eating habits of the baby to be born; and negotiating with partners about the division of domestic labour. Enhanced Triple P for Baby participants were less likely, compared to MB attendees, to say that they felt social interaction between group members, in and of itself, provided any special benefits. However, a few ETPB groups, smaller than recommended due to low or patchy attendance, appeared to facilitate more in-depth discussion than was typical. Members of one such group identified, during their discussions, that they had anxiety issues in common. They continued meeting regularly, after the antenatal sessions ended. How well do participants like the programme, why/why not, do they think it is useful and in what ways? Mellow Bumps: Based on evaluation forms completed at the end of each session, those who ‘strongly agreed’ or ‘agreed’ that they learnt a lot ranged from 60% (51/84) for session 1 (putting participants at their ease; establish safe environment; dispel false myths about motherhood; normalise reactions to babies if these are not completely positive; introduce capacities of babies; reduce maternal stress) to 90% for session 6 (53/59; plan future support for mother and baby). The percentage of participants who ‘strongly agreed’ or ‘agreed’ that the materials were useful ranged from 85% (71/84) for session 1 to 97% for session 3 (61/63; addressing past interferences or support for being a parent). Only three groups included an optional session with partners, and of these 4/8 of participants ‘strongly agreed’ or ‘agreed’ that they learnt a lot and 6/8 of participants ‘strongly agreed’ or ‘agreed’ that the materials were useful. For the programme as a whole, based on post-intervention evaluations completed after the last antenatal session, the percentage of participants who ‘strongly agreed’ or ‘agreed’ that the sessions had been helpful was 82% (41/50), informative was 78% (38/49) and enjoyable was 100% (50/50). The aspects rated most highly (i.e. by the largest percentage of participants) were ‘tips on managing stress’ and ‘awareness of how mothers should act with their baby’, both of which 86% (44/51) of participants ‘strongly agreed’ or ‘agreed’ group attendance had furnished them with. In addition, 84% (43/51) of participants ‘strongly agreed’ or ‘agreed’ that the groups had helped make them more confident about being a mother, helped make them feel more confident in talking to others and had given them a greater understanding of what a baby needs. The group context rated highly in terms of what the participants liked most about the sessions, with the sharing of experiences and thoughts valued in particular. Indeed, in rating the extent they felt listened to in each session, the proportion of participants ‘strongly agreeing’ or ‘agreeing’ that they felt listened to ranged from 91% (50/55) for session 5 (the outing or activity) to 98% (57/58) for session 6. Among the eight women who attended an optional partners session, 100% felt listened to during this session. Very small numbers of women said that they felt left out in any session (0 women reported this for session 1 or the partner’s session; 1 woman for sessions 3 and 4; 2 for session 2; 3 for session 5 and 4 for session 6). Enhanced Triple P for Baby: Based on evaluation forms completed at the end of each session, those who ‘strongly agreed’ or ‘agreed’ that they learnt a lot ranged from 84% (73/87) for session 1 (positive parenting and baby development) to 97% for session 3 (67/69, survival skills). The percentage of participants who ‘strongly agreed’ or ‘agreed’ that the materials were useful ranged from 91% (79/87) for session 1 to 97% for session 3 (67/69) and session 4 (60/62, partner support). Learning rated highly in terms of what the participants liked most about the sessions, though in session 4 it was sharing and discussion that was most valued (followed by learning), and that group context was valued overall. Most participants felt they had been listened to in each session, with the proportion of participants ‘strongly agreeing’ or ‘agreeing’ that they felt listened to ranging from 97% (84/87) for session 1 and session 4 (60/62) to 100% for session 3. Only a very small number of women said that they felt left out in any session: four women reported this for session 1 and session 4, one woman for session 3 and no one for session 2 (responding to your baby). For the programme as whole, based on post-intervention evaluations completed after the last antenatal session, 98% (58/59) ‘strongly agreed’ or ‘agreed’ that the sessions were helpful, 98% (58/59) that they were informative, and 100% (62/62) that they were enjoyable. The aspects rated most highly (i.e. by the largest percentage of participants) were ‘tips on managing stress’ and ‘practical tips on caring for baby’, which 97% (61/63) of participants ‘strongly agreed’ or ‘agreed’ group attendance had furnished them with, and 95% (60/63) ‘agreed’ or ‘strongly’ agreed the sessions had helped with ‘awareness of how they should act with the baby’ and 94% (59/63) with ‘understanding of what a baby needs’. In accordance with the different theories of change of the two programmes, a higher percentage of ETPB participants reported the programme informative compared with MB participants, and a higher percentage reported learning a lot. For both programmes, 100% of participants found the sessions enjoyable and most women valued the group aspect and felt that they were listened to. Does partner’s support affect experience of programme? There seems to be little data relating to whether partners’ support affected participants’ experience of the programme, but there are data relating to the experience of the programme and partners’ support, with both positive accounts ( ETPB participants) and accounts lamenting that the programmes were not more influential in this respect ( MB and ETPB). Generally, ETPB was viewed as valuable in this respect. Do other potential sources of support, for example parents, siblings and friendship networks, affect experience of programme? It was striking among the MB participants that many reported little other support. Perhaps MB has the potential to be most effective for those without a support network, though there is the question of how lasting connections made during a seven-session (or less) programme can be. Future analyses should explore this question. As discussed above, we do have data on the extent to which participants kept in touch beyond the programme, but this did not seem to happen often due to the nature of the trial and preventing people ‘timing out’ (i.e. making sure that participants received the antenatal component of the intervention during pregnancy), there was a tendency for participants to have to travel out of their area. How does composition of the group in terms of different additional needs affect group dynamics? Regarding the reflective work in MB a minority of participants were marginalised by others in the group, even when group dynamics among remaining members were positive. 119 The social nature of the group meant that participants with complex health and social care problems felt they had to protect others, for example, those who might not have been exposed to drug use, child removals, etc. One woman who participated in MB said she wanted to avoid ‘ruining’ the experiences of first-time mothers-to-be in her group, so she stayed quiet about her own circumstances. For MB this may have had a negative impact on the nurturing ethos of the group. Do participants think the facilitators understand them? Participants generally seemed to like the facilitators, feeling warm towards them. Session evaluation forms suggest that nearly all of the participants felt listened to in the group, for both ETPB and MB rates were very high across sessions. Would participants trust facilitators with sensitive information? Most participants trusted their facilitator(s) and other group members with sensitive information. However, a minority of participants said that they chose not to disclose anything they felt was personal, usually because they did not feel comfortable in that group. According to the programme theory for ETPB , it was less likely that sensitive information would be disclosed in a group context. However, facilitator reports suggest that one-to-one contact sometimes created a different dynamic where personal circumstances might be explored in depth and there was sometimes scope, in smaller groups, for in-depth discussions. There did appear to be trust between participants and facilitators generally, and questions of a sensitive nature could be asked. To what extent did care-as-usual alone meet participants’ needs? Antenatal support, in the context of additional care and support needs, typically involved being assigned a SNIP (NHS GGC) or Vulnerable in Pregnancy (VIP) (NHS A&A) midwife and being referred to other services (e.g. drug and alcohol services, perinatal mental health services, etc) as considered appropriate. CAU participants were also referred, as standard, to individual or group-based interventions if available in the area (e.g. Barnardos, Threads, Breathing Space, FNP , etc. see Appendix 6 , Table 22 ). Concerns had been expressed by some RPs that CAU participants might be ‘left with no intervention at all’. We wonder if this might have had the unintended consequence of increasing the number of referrals for this group, although we are unable to say. However, we can note that multiple referrals to a range of services/interventions did not necessarily result in participants’ needs being met. Most participants indicated that at least one critical social or healthcare issue was addressed well but that other services/interventions were not satisfactory. Disentangling precisely what worked or not for whom, and why, for CAU participants was difficult because of the mix of problems they had and the different combination of interventions that might have been offered to them. Special Needs in Pregnancy and the VIPs appeared to provide continuity of relationships with midwives for most participants due to having more meetings than would be the case for women without additional social and care needs. This was perceived to be an advantage over mainstream midwifery care, and particularly valuable for pregnant women with special needs, as it avoided the need to repeat complex personal and medical histories with different practitioners. The majority of CAU participants perceived SNIP/VIP services to be sensitive and working effectively to reduce stigma. This appeared to make it acceptable, for most, to have their medical, social and psychological problems dealt with together. Confidence appeared to be high that problems would be dealt with sympathetically and that knowledge of one issue would not negatively impact the handling of another. One exception, a participant in her thirties with a history of alcohol problems who had been homeless at the point of interview, said she felt she had been treated like she was ‘lower down the social scale’ at the SNIP clinic. She had overheard a member of staff ‘moaning about … the homeless people that came in off the street’ (M19 GGC). The language used was stigmatising and upset her. This participant reverted back to mainstream care to avoid others’ perceptions of her history of substance misuse/homelessness from contaminating her experience of pregnancy. She sought help for her alcohol problem separately. Several participants highlighted how little time they felt there was in midwifery appointments (SNIP/VIP or ‘mainstream’) to discuss their emotions. All focus was perceived to be on the physical health of the mother and well-being of the baby at the expense of the mother’s mental health, and participants were acutely aware of how busy MWs were. One participant thought that most of the things that helped during her pregnancy [National Childbirth Trust (NCT) classes, pregnancy yoga, aqua yoga] and after birth (a private appointment with a sleep consultant, baby sensory classes, mother and baby yoga) she had had to arrange and pay for herself. She felt fortunate that she had the material and personal resources to do this to help manage her poor mental health. However, she was dissatisfied with the lack of ‘proper’ treatment offered when she had sought emotional support through her MW and GP . Others who considered they had had ‘atypical’ experiences during pregnancy (e.g. hyperemesis gravidarum, news of fetal abnormality, twin pregnancies) also did not feel well supported emotionally. While some had the personal resources to seek out and obtain emotional support, for example through social media, others felt bewildered and did not know who to ask for help. One example of the latter was a participant with a learning disability who had recently learnt her baby had a hole in his heart. Practical support for broader social problems was lacking in ETPB and MB , although observations revealed that MB at least gave participants the space to discuss anxieties about housing and managing on low incomes, if not the practical support to obtain resolutions. Care-as-usual appeared better at supporting some participants who felt stressed by their social problems. The youngest participant (T1), a teenager with concerns about lack of money and housing, was helped by her family nurse partner to apply for housing and benefits. Other CAU participants were offered practical support when they needed it postnatally, such as being picked up (by a SNIP MW or the group facilitator) from home and taken to weekly sessions at a community-based group. This appeared to reduce anxiety about group attendance and help these participants engage in activities that had potential benefits for both mother and baby. The study was not narratively informed that CAU were purposively being offered more support but there is some evidence of it from the EE data ( Table 22 ). Complementary PhD studies linked to THRIVE The MRC funded two complementary PhD studentships. The first was conducted by Karen Maxwell and supervised by Katie Buston and Daniel Wight to study the partners of mothers recruited to the trial. 38 The project explored: (a) how mothers’ partners influenced mothers’ participation in the trial; and (b) whether the formal involvement of fathers in ETPB was critical to the outcomes of the programme. The thesis, completed in 2018, contributes substantially to our overall understanding of the experiences of marginalised fathers. Maxwell’s 38 conclusions were that the fathers’ of the THRIVE babies held normative ideals about good fathering, but that they faced challenge in enacting these ideals. To target disadvantaged fathers more effectively, Maxwell 38 suggested interventions should, ‘capitalise on men’s excitement and commitment to partner and baby in the antenatal period; emphasize the relevance of content to the needs of disadvantaged men; and bear in mind potential barriers such as perceived lack of “need”, overcoming social anxieties, and notions of acceptable masculinity’. 38 The second MRC-funded PhD studentship was conducted by Simon Barrett, supervised by Katie Buston and Marion Henderson, and was entitled ‘How people’s own upbringing influences their subsequent parenting practices: a mixed method study of mothers and/or their partners with additional health and social care needs’. 39 The thesis, completed in 2020, explored how parents’ own childhood experiences affect parenting behaviour, sampling mothers from the THRIVE study for in-depth interviews. It also examines the relationships between adverse childhood events and parental self-efficacy through analyses of the THRIVE outcome data set. The conclusions of this theses were that consciously or unconsciously parenting practices were transmitted across generations. Deprivation and histories that include trauma, can challenge ‘mothers to consistently parent in warm and supportive ways and break unwanted cycles. …Interventions targeted at these mothers should: incorporate ways of alleviating the added difficulties associated with these contextual factors; seek to improve maternal self-efficacy; and take account of the subjective nature of parenting and internalised norms and beliefs’. 39 Conclusions Key findings from the process evaluation are summarised within the Discussion in the context of the outcome evaluation and EE . Our main conclusion is that, on the basis of the PE data, we anticipated in advance that there would be limited evidence of effectiveness revealed by the outcome evaluation. Enhanced Triple P for Baby appeared to be easier to implement as theorised compared with MB , at least during the antenatal phase, but fidelity across both interventions was by no means perfect and attendance was concerning. Copyright © 2025 Henderson et al . This work was produced by Henderson et al . under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/ . For attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited. Bookshelf ID: NBK614343 Contents < Prev Next > Share Views PubReader Print View Cite this Page Henderson M, Wittkowski A, Buston K, et al. Evaluation of parenting interventions for those with additional health and social care needs during pregnancy: THRIVE a multi-arm RCT with embedded economic and process components. Southampton (UK): National Institute for Health and Care Research; 2025 May. (Public Health Research, No. 13.04.) Chapter 6, Process evaluation. PDF version of this title (3.3M) In this Page Introduction Findings Complementary PhD studies linked to THRIVE Conclusions Other titles in this collection Public Health Research Recent Activity Clear Turn Off Turn On Process evaluation - Evaluation of parenting interventions for those with additi... Process evaluation - Evaluation of parenting interventions for those with additional health and social care needs during pregnancy: THRIVE a multi-arm RCT with embedded economic and process components Your browsing activity is empty. Activity recording is turned off. Turn recording back on See more... Follow NCBI Twitter Facebook LinkedIn GitHub NCBI Insights Blog Connect with NLM Twitter Facebook Youtube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov