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
Background Women who have additional social and care needs in pregnancy (e.g. social adversity, maternal depression and anxiety) are likely to produce high levels of stress hormones. This has the potential to affect fetal brain development, increase infant reactivity to stress, and impair sensitive...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
NIHR Journals Library
2025-05-01
|
| Series: | Public Health Research |
| Subjects: | |
| Online Access: | https://doi.org/10.3310/KYMT5407 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850035132021342208 |
|---|---|
| author | Marion Henderson Anja Wittkowski Katie Buston Karen Crawford Alice MacLachlan Alex McConnachie Emma McIntosh Claudia-Martina Messow Catherine Nixon Rosaleen O’Brien Shona Shinwell Daniel Wight Yiqiao Xin Rachel Calam Ruth Dundas James Law Helen Minnis Lucy Thompson Philip Wilson |
| author_facet | Marion Henderson Anja Wittkowski Katie Buston Karen Crawford Alice MacLachlan Alex McConnachie Emma McIntosh Claudia-Martina Messow Catherine Nixon Rosaleen O’Brien Shona Shinwell Daniel Wight Yiqiao Xin Rachel Calam Ruth Dundas James Law Helen Minnis Lucy Thompson Philip Wilson |
| author_sort | Marion Henderson |
| collection | DOAJ |
| description | Background Women who have additional social and care needs in pregnancy (e.g. social adversity, maternal depression and anxiety) are likely to produce high levels of stress hormones. This has the potential to affect fetal brain development, increase infant reactivity to stress, and impair sensitive mother–infant bonds from developing. These in turn may have long-term effects on children’s health, social and educational outcomes. Parenting interventions show promising improvements to child outcomes; however, there is little evidence of their efficacy in the UK. Objective(s) THRIVE compared the impact of taking part in one of two antenatal parenting support programmes both incorporating cognitive–behavioural therapy (Enhanced Triple P for Baby or Mellow Bumps) with care-as-usual alone on the mental health and maternal attunement of vulnerable mothers-to-be, as well as the socioemotional and behavioural development of their children. Design THRIVE is a three-arm randomised controlled trial. Pregnant women with additional social and care needs in pregnancies were invited to participate. Participants were randomly allocated to Enhanced Triple P for Baby, Mellow Bumps or care-as-usual Setting The study took place in National Health Service Greater Glasgow and Clyde and National Health Service Ayrshire and Arran health board areas. Intervention sessions were predominantly in community settings. Participants Women identified as having additional social and care needs in pregnancies during pregnancy based on the National Health Service Greater Glasgow and Clyde’s Special Needs in Pregnancy criteria were recruited to THRIVE (n = 485), slightly below the target of 500 women. Participants were block-randomised 5 : 5 : 2 to Enhanced Triple P for Baby, Mellow Bumps or care-as-usual. Subsequently their babies and accompanying person of choice were also invited. Interventions Enhanced Triple P for Baby consists of four weekly group-based antenatal sessions followed by up to three postnatal home visits and one postnatal group session. It aims to provide babies with a healthy start to life by combining parenting skills training with strategies to enhance individual well-being and couple adjustment. Mellow Bumps in comparison, comprises of seven weekly antenatal sessions and one postnatal session. It aims to decrease maternal stress, increase understanding of neonates’ capacity for social interaction, and emphasise the importance of early interaction for brain development and attachment. Recruitment to the trial took place between early 2014 and May 2018. Main outcome measures Our two primary outcomes were the Hospital Anxiety and Depression Scale, plus outwardly directed irritability from the Adult Wellbeing Scale, and CARE Index mother–infant dyadic interaction synchrony. Results There were no significant differences in the change from baseline in Hospital Anxiety and Depression Scale and outwardly expressed irritability score between the combined active intervention groups and care-as-usual only [effect size (95% confidence interval) 0.03 (−0.24 to 0.29)], or between either Enhanced Triple P for Baby and Care-as-usual only [0.04 (−0.24 to 0.32)] or Mellow Bumps and care-as-usual only [0.01 (−0.27 to 0.30)]. Mellow Bumps holds promise for cost-effectiveness due to its potential for cost-savings relating to routine healthcare resource use. Limitations The main limitation of THRIVE was poor attendance of the groups (under 42%) offered to our Enhanced Triple P for Baby and Mellow Bumps participants. However, subgroup analysis adjusted for level of attendance suggested the results would not have been any different with increased participation. Conclusions With the population THRIVE involved, at a stage of pregnancy when a lot of support is offered (with many parenting interventions being offered without rigorous evaluation), we do not have evidence to recommend the rollout of Enhanced Triple P for Baby or Mellow Bumps. Future work Future work could explore the timing and intensity of interventions, intervention group composition, co-produced programmes, online sessions, and impact of addressing inequalities. Trial registration This trial is registered as Current Controlled Trials ISRCTN21656568 (www.isrctn.com/ISRCTN21656568). Funding This award was funded by the National Institute for Health and Care Research (NIHR) Public Health Research programme (NIHR award ref. 11/3002/01) and is published in full in Public Health Research; Vol. 13, No. 4. See the NIHR Funding and Awards website for further award information.
Plain language summary What was the question? Women experiencing high levels of adversity, maternal depression and/or anxiety during pregnancy are more likely to feel stressed. That stress may lead to their baby being less able to cope with stress and may hamper the sensitive mother–baby relationship from developing. The THRIVE study aimed to explore ways that may help mothers protect their mental health and relationship with their babies. What did we do? Women were invited to take part in the THRIVE study by midwives and researchers. THRIVE took place in National Health Service Greater Glasgow and Clyde and National Health Service Ayrshire and Arran. The women who agreed to take part were randomly selected into 1 of 3 groupings. Group 1 received Enhanced Triple P for Baby plus care-as-usual, Group 2 received Mellow Bumps plus care-as-usual and Group 3 continued with care-as-usual only. We had two main aims. The first was to see if mothers’ mental health (anxiety and depression and reduce outwardly directed irritability) was better if they received either Enhanced Triple P for Baby or Mellow Bumps (interventions) compared with care-as-usual only. The second was to explore if the observed quality of relationship between mother and baby was stronger if they received an intervention versus care-as-usual only. What did we find? Our results showed that neither intervention was successful in improving either mental health or the observed quality of mother–baby interactions. Mellow Bumps is most likely to be deemed cost-effective due to its potential for cost-savings arising in routine healthcare resource use. What does this mean? We cannot recommend Enhanced Triple P for Baby or Mellow Bumps as being better than care-as-usual in the UK. We recommend that future work could explore the timing and intensity of interventions, intervention group composition, online sessions, co-produced programmes, and the impact of addressing inequalities.
Scientific summary Background Evidence is growing that depression, stress, and anxiety in pregnant women can create adverse modifications to the fetus in utero that can permanently negatively influence the baby’s response to stress and disrupt the mother’s ability to be sensitive to her baby. As poor mother–baby interaction and maternal mental ill health often lead to child maltreatment, most postnatal interventions may not be able to undo some of the damage sustained by babies due to their mother’s and/or parents’ response to adverse circumstances during pregnancy. Women additional social and care needs in pregnancy (ASCNs e.g. mental ill health, domestic abuse, care experience) are likely to be more anxious, depressed and produce higher levels of stress-related hormones than women in more favourable circumstances. The Trial for Healthy Relationship Initiatives in the Very Early Years (THRIVE) study evaluated the effects of two parenting interventions on maternal mental health and mother–infant interactions for women who are identified as having ASCNs. Each intervention was provided in addition to care-as-usual (CAU), Enhanced Triple P for Baby (ETPB) and Mellow Bumps (MB) versus CAU. Objectives The overall aim of THRIVE is to evaluate the impact of the ETPB + CAU (ETPB) and MB + CAU (MB) parenting interventions on maternal mental health and mother–child relationships, when compared with CAU alone. In the longer term, THRIVE will also assess whether children whose mothers received ETPB or MB show reduced incidence of child maltreatment. Due to substantial amendments, our secondary research questions changed from protocol, mostly related to us not collecting data when the babies reached 18 months. The primary research questions that are addressed in this report were: Do mothers receiving ETPB or MB show significantly: Lower anxiety, depression and outwardly directed irritability compared with those receiving CAU only when their babies are about 6 months old? Higher maternal sensitivity scores in mother–infant interaction compared with those receiving CAU when their babies are about 6 months old? Secondary research questions included: Do infants whose parents receive ETPB or MB show more co-operative behaviour signs than those whose parents received CAU? Are either ETPB or MB more effective for different subgroups of parents? Are either ETPB or MB cost-effective for the NHS or society more broadly, in the long term? Do differences in programme fidelity, practitioners’ characteristics and motivation, mothers’ engagement, intervention mechanisms, and contextual factors affect mother and infant outcomes? Methods Study design A three-arm randomised controlled trial (RCT) comparing two parenting interventions, ETPB and MB, with CAU only. The study included women who received their care in NHS Greater Glasgow and Clyde (GGC) or NHS Ayrshire and Arran (A&A). Participants Women identified as having ASCNs during pregnancy based on the NHS GGC’s Special Needs in Pregnancy criteria were recruited to THRIVE. Subsequently, their babies and partner/accompanying person of choice were also invited to be involved in THRIVE. Interventions Both interventions aim to reduce maltreatment and improve sociodevelopmental outcomes for children, with content including information on infant development and incorporating cognitive–behavioural therapy to improve mothers’ coping in a general and parenting context. The set-up of the two interventions is also similar in terms of working with women from 20 weeks of pregnancy in groups of about eight participants with sessions delivered in local community venues by two trained practitioners. However, they have fundamental differences in focus and mechanism that may influence how effective they are at reducing child maltreatment. ETPB incorporates social learning principles, with an emphasis on families. ETPB includes fathers or an accompanying person of the participants’ choice and includes skills-based content focused on expectations and coping to meet the new challenges of parenthood while maintaining a happy family. MB is underpinned by attachment theory and designed to target mothers who are vulnerable in pregnancy. The focus of MB is on mothers, although fathers can be invited to one session, and the content focuses on nurturing mothers’ self-care, providing guided reflection, encouraging nurturing of the fetus/baby, engagement with the fetus/baby and synchrony in the mother–fetus/infant relationship. Primary study outcomes The following primary study end points were assessed between 6 and 12 months post partum using validated scales: Maternal mental health measured using the Hospital Anxiety and Depression Scale (HADS) and outwardly expressed irritability (I) measures from the Adult Wellbeing Scale (AWS). The dyadic synchrony between the dyad within the mother–baby interaction was measured using the CARE Index. The CARE Index uses video recordings of 3–5 minutes of mother–baby play interaction, which were blind coded by accredited coders. Secondary study outcomes The following secondary outcomes were assessed between 6 and 12 months postnatal: HADS depression subscale. HADS anxiety subscale. AWS outwardly expressed irritability subscale. CARE Index co-operative behaviour. The intervention group was entered in the regression models in the following ways: Binary as active intervention (ETPB or MB) versus CAU. Binary as ETPB versus MB (on subgroup receiving any active intervention). Categorical with the three categories ETPB, MB and CAU. Economic evaluation outcomes EuroQol-5 Dimensions (EQ-5D) total domain score and visual analogue scale (VAS). Costs were adjusted using generalised linear models with gamma family and log link to account for covariates and skew distribution of cost. Then incremental cost-effectiveness ratio (ICER) in the cost–utility analysis was calculated as the ratio of incremental cost difference and incremental quality-adjusted life-years (QALYs) difference (ETPB vs. CAU and MB vs. CAU). Process evaluation A realist process evaluation (PE) was undertaken and utilised a range of methods, including participant observation, interviews and questionnaires to examine ‘what works for whom in what circumstances?’ The main themes were independently coded across a sample of interviews by two researchers, Rosaleen O’Brien (RO) and Katie Buston (KB), using NVivo 10 (QSR International, Warrington, UK). A coding framework was then agreed by RO and KB and discussed/refined with the THRIVE PE subgroup, based on the key questions for the PE, developed prior to implementation, with the addition of novel areas identified through data collection. Recruitment data collection and analysis Participants were recruited to THRIVE over 53 months from January 2014 to May 2018. Three main recruitment strategies were used: referral from health or social care practitioner or third-sector organisation, who screened for eligibility and interest for participation in the trial from October 2015 (following a protocol amendment), participants could be recruited in person in a clinic or community setting by a member of the THRIVE team; and self-referral to the trial by responding to poster advertisements in community/healthcare setting, targeted web-based advertisement, or social media advertising. For strategies 2 and 3, potential participants were screened for eligibility and a referral form was completed by THRIVE staff. Following receipt of a referral form, or a note of, details of potential participants were entered in the study database and the NHS GGC Clinical Research Facility were contacted to determine whether the pregnancy was continuing. Once confirmed, a member of the THRIVE team contacted potential participants to arrange an appointment at which they were afforded the opportunity to ask questions about the research, and on agreeing to participate, were consented to trial and had baseline measures completed. Results From the 964 referrals received, 488 participants were recruited to THRIVE (marginally below the aspired 500) and completed baseline assessments. The intention-to-treat (ITT) population included 450 participants. Among the ITT population, 135 participants who received ETPB, 131 who received MB and 53 who received CAU only, completed follow-up measures at 6–12 months post partum. We therefore reached statistical power for our primary and secondary analyses. The demographic and baseline characteristics of THRIVE participants were balanced across groups. The mean age was 26.7 years, 63.0% of participants were in the Scottish Index of Multiple Deprivation quintile 1 (i.e. the 20% most deprived areas) and 94.4% reported white ethnicity (the 2011 Scottish Census showed that 96% reported a white ethnicity). This was the first reported pregnancy for 45.1% of participants, and 40.2% of all participants reported that their pregnancy was planned by both parents. Primary study outcomes Hospital Anxiety and Depression Scale plus irritability At baseline, the mean standard deviation (SD) HADS + I score was 0.31 (0.17) for ETPB, 0.31 (0.17) for MB and 0.32 (0.17) for CAU only. Mean changes from baseline to follow-up (6–12 months post partum) were −0.03 (0.18) for ETPB, −0.03 (0.16) for MB and −0.04 (0.15) for CAU only. These mean scores suggest that on average the participants were not depressed, anxious, or irritable at the start of the trial or at 6–12 months post partum. There were no significant differences in the change from baseline in the HADS + I score between the combined active intervention groups and CAU only {effect size [95% confidence interval (CI)] 0.03 [−0.24 to 0.29]}, or between either ETPB and CAU only [0.04 (−0.24 to 0.32)] or MB and CAU only [0.01 (−0.27 to 0.30)]. The same pattern of results was found in the per-protocol (PP) analysis and in analyses conducted to assess the sensitivity of HADS + I to missing values and clustering within intervention groups. Mother–infant dyadic interaction synchrony The mother–infant dyadic interaction synchrony was assessed at follow-up when babies were 6–12 months old using the CARE Index dyadic synchrony score. Mean (SD) dyadic synchrony scores were 6.6 (2.7) for ETPB, 6.4 (2.5) for MB and 7.1 (2.6) for CAU. These scores suggest the mother–infant scores in THRIVE were slightly below average in terms of maternal sensitivity and baby’s co-operativeness. There were no significant differences in the CARE Index dyadic synchrony score between the combined active intervention groups and CAU only [effect size (95% CI) −0.20 (−0.52 to 0.12)], or between either ETPB and CAU only [−0.17 (−0.52 to 0.18)] or MB and CAU only [−0.24 (−0.59 to 0.11)]. The same pattern of results was found in the PP analysis, analyses conducted to assess the sensitivity of HADS + I to missing values and clustering within intervention groups and after removing dyadic synchrony scores generated from episodes of caregiving rather than play. Neither mediator and moderator nor subgroup analysis was able to show any significant results. Secondary study outcomes When the HADS + I score was broken down into its subscales, there were no significant differences between treatment groups for HADS depression score, HADS anxiety score, or outwardly expressed irritability score. There were also no significant differences between groups for other CARE Index outcomes, including the mother’s sensitivity score or the co-operative behaviour score. Several other secondary outcomes were assessed, including support and control in childbirth, post-partum partner support, parental regulation, emotional regulation, maternal attitudes to marital relationships, perceived social support, parenting self-confidence, maternal attitudes, parental coping, response to infant crying and post-partum bonding. The only secondary outcome for which significant differences were observed between treatment groups was the Cognitive Emotional Regulation Questionnaire. Significantly lower scores were observed for ETPB versus CAU only for the focus on thought subscale (better) and the focus on planning subscale (worse) and for MB versus CAU only for the acceptance subscale (worse), the focus on thought subscale (better) and the positive reappraisal subscale (worse). Economic evaluation outcomes Quality-adjusted life-years Participants had lower utility at follow-up one for both ETPB and MB groups compared to the CAU arm; the adjusted mean utilities at ETPB group and MB group were 0.011 (95% −0.079 to 0.063) and 0.021 (95% −0.095 to 0.061) lower than CAU. There was no statistically significant difference between QALYs gained in each group over the 1-year period. Cost–utility results Compared to the CAU, both ETPB and MB group had a decreased point estimate of cost, at −£703 (−£3725, £2045) and −£1570 (−£4487, £703), respectively, which demonstrated a potential for cost-saving. Please note, this is not cost-effective in the traditional sense, as there were no quality-of-life savings, but in terms of cost-savings and against a background of great uncertainty in the effectiveness outcome. When hospitalisation costs were excluded in the sensitivity analysis, MB was no longer a cost-saving and CAU became the most cost-effective option due to its lower cost. Both ETPB and MB groups did not show any QALY gain compared to the CAU group; both with a slight, statistically insignificant, QALY loss of 0.23 over the 1-year period. Given both the mean incremental cost and mean QALY gain were negative, net monetary benefit was used instead of the ICER to determine the cost-effectiveness against the threshold £20,000. It shows that MB had the highest point estimate of the net monetary benefit, at £6035 (95% CI £4120 to £7800), followed by ETPB at £5167 (95% CI £2551 to £6965) and CAU at £4905 (95% CI £1960 to £7051). At £20,000 per QALY threshold, the probability for MB to be cost-effective was highest, at 65%, and 22% for ETPB, and 13% for CAU. Process evaluation The main conclusion, anticipated in advance, was that there would be limited evidence of effectiveness revealed by the outcome evaluation. ETPB appeared to be easier to implement as theorised, at least during the antenatal phase, but fidelity was by no means perfect, and attendance was concerning for both interventions. Limitations The main limitation of THRIVE was poor attendance at the groups (under 42%) offered to the ETPB and MB participants. However, a subgroup analysis that adjusted for level of attendance did not suggest that the results would have been different if participants had taken up more of the interventions. Conclusions THRIVE was not able to demonstrate an impact of ETPB or MB on prespecified primary outcomes among participants with additional social and care needs, compared with CAU. The economic evaluation suggests that MB may help reduce routine health care, but this finding needs to be interpreted with caution. There is clearly much need for support among the THRIVE population, and CAU works intensively with families at this stage of life, it may be that ETPB or MB was not able to add much more value to the CAU being offered. We suggest working with those with lived experience of additional social and care needs to explore other avenues of support that can be complementary to CAU. Our recommendations for future action and research, in numbered order, are to explore: Timing and intensity More research is needed to identify the best timing for early years parenting interventions, to complement CAU, and ideally programmes that can adapt to the intensity required for parents with different levels of additional social and care needs. Group composition It would be interesting to explore the benefit of more homogeneous groups’ composition rather than the artefactually heterogeneous groups that were evaluated in THRIVE. Complementary app/online sessions Some women could not attend group sessions due to work, child care or ill-health issues, but may have engaged with online information. So, exploring hybrid engagement opportunities may be useful in the future. Women with mental ill health or social anxieties may also prefer online options. Co-produced programmes Enhanced Triple P for Baby and MB were developed by experts in clinical psychology for participants with additional care and support needs but they may not be experts in the THRIVE population. A higher level of involvement with people with lived experience in the development of interventions may help programmes to be delivered when most beneficial to potential participants. It is likely that there is no one size fits all and that support for those with additional social and care needs may need to be able to respond flexibly to bespoke and complex life situations. Exploring ways to support communities to help themselves may be beneficial. A literature review of co-production in mental health suggests that this is a promising approach. Addressing inequalities There is a large body of evidence showing the impact of inequalities on health. Moving towards a more egalitarian society may lead to less pressure on the NHS and need for programmes, such as ETPB and MB. Trial registration This trial is registered as Current Controlled Trials ISRCTN21656568. Funding This award was funded by the National Institute for Health and Care Research (NIHR) Public Health Research programme (NIHR award ref. 11/3002/01) and is published in full in Public Health Research; Vol. 13, No. 4. See the NIHR Funding and Awards website for further award information. |
| format | Article |
| id | doaj-art-1b0f9002b99b483296d306635558b7d8 |
| institution | DOAJ |
| issn | 2050-439X |
| language | English |
| publishDate | 2025-05-01 |
| publisher | NIHR Journals Library |
| record_format | Article |
| series | Public Health Research |
| spelling | doaj-art-1b0f9002b99b483296d306635558b7d82025-08-20T02:57:34ZengNIHR Journals LibraryPublic Health Research2050-439X2025-05-01130410.3310/KYMT540711/3002/01Evaluation of parenting interventions for those with additional health and social care needs during pregnancy: THRIVE a multi-arm RCT with embedded economic and process componentsMarion Henderson0Anja Wittkowski1Katie Buston2Karen Crawford3Alice MacLachlan4Alex McConnachie5Emma McIntosh6Claudia-Martina Messow7Catherine Nixon8Rosaleen O’Brien9Shona Shinwell10Daniel Wight11Yiqiao Xin12Rachel Calam13Ruth Dundas14James Law15Helen Minnis16Lucy Thompson17Philip Wilson18Medical Research Council/Chief Scientist Office Social and Public Health Sciences Unit, University of Glasgow, Glasgow, ScotlandDivision of Psychology and Mental Health, School of Health Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester, EnglandMedical Research Council/Chief Scientist Office Social and Public Health Sciences Unit, University of Glasgow, Glasgow, ScotlandRobertson Centre for Biostatistics, University of Glasgow, Glasgow, ScotlandMedical Research Council/Chief Scientist Office Social and Public Health Sciences Unit, University of Glasgow, Glasgow, ScotlandRobertson Centre for Biostatistics, University of Glasgow, Glasgow, ScotlandHealth Economics and Health Technology Assessment, University of Glasgow, Glasgow, ScotlandRobertson Centre for Biostatistics, University of Glasgow, Glasgow, ScotlandMedical Research Council/Chief Scientist Office Social and Public Health Sciences Unit, University of Glasgow, Glasgow, ScotlandParenting and Family Support Research Programme, Department of Psychology and Allied Health Sciences, School of Health and Life Sciences, Glasgow Caledonian University, Glasgow, ScotlandMother and Infant Research Unit, School of Nursing and Health Sciences, University of Dundee, Nethergate, Dundee, ScotlandMedical Research Council/Chief Scientist Office Social and Public Health Sciences Unit, University of Glasgow, Glasgow, ScotlandHealth Economics and Health Technology Assessment, University of Glasgow, Glasgow, ScotlandDivision of Psychology and Mental Health, School of Health Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester, EnglandMedical Research Council/Chief Scientist Office Social and Public Health Sciences Unit, University of Glasgow, Glasgow, ScotlandInstitute of Health and Society, School of Education, Communication and Language Sciences, University of Newcastle, Newcastle-upon-Tyne, EnglandInstitute of Health and Wellbeing, University of Glasgow, Level 4, Academic CAMHS, Yorkhill Hospital, Glasgow, ScotlandInstitute of Health and Wellbeing, University of Glasgow, Level 4, Academic CAMHS, Yorkhill Hospital, Glasgow, ScotlandCentre for Rural Health, University of Aberdeen, The Centre for Health Science, Inverness, ScotlandBackground Women who have additional social and care needs in pregnancy (e.g. social adversity, maternal depression and anxiety) are likely to produce high levels of stress hormones. This has the potential to affect fetal brain development, increase infant reactivity to stress, and impair sensitive mother–infant bonds from developing. These in turn may have long-term effects on children’s health, social and educational outcomes. Parenting interventions show promising improvements to child outcomes; however, there is little evidence of their efficacy in the UK. Objective(s) THRIVE compared the impact of taking part in one of two antenatal parenting support programmes both incorporating cognitive–behavioural therapy (Enhanced Triple P for Baby or Mellow Bumps) with care-as-usual alone on the mental health and maternal attunement of vulnerable mothers-to-be, as well as the socioemotional and behavioural development of their children. Design THRIVE is a three-arm randomised controlled trial. Pregnant women with additional social and care needs in pregnancies were invited to participate. Participants were randomly allocated to Enhanced Triple P for Baby, Mellow Bumps or care-as-usual Setting The study took place in National Health Service Greater Glasgow and Clyde and National Health Service Ayrshire and Arran health board areas. Intervention sessions were predominantly in community settings. Participants Women identified as having additional social and care needs in pregnancies during pregnancy based on the National Health Service Greater Glasgow and Clyde’s Special Needs in Pregnancy criteria were recruited to THRIVE (n = 485), slightly below the target of 500 women. Participants were block-randomised 5 : 5 : 2 to Enhanced Triple P for Baby, Mellow Bumps or care-as-usual. Subsequently their babies and accompanying person of choice were also invited. Interventions Enhanced Triple P for Baby consists of four weekly group-based antenatal sessions followed by up to three postnatal home visits and one postnatal group session. It aims to provide babies with a healthy start to life by combining parenting skills training with strategies to enhance individual well-being and couple adjustment. Mellow Bumps in comparison, comprises of seven weekly antenatal sessions and one postnatal session. It aims to decrease maternal stress, increase understanding of neonates’ capacity for social interaction, and emphasise the importance of early interaction for brain development and attachment. Recruitment to the trial took place between early 2014 and May 2018. Main outcome measures Our two primary outcomes were the Hospital Anxiety and Depression Scale, plus outwardly directed irritability from the Adult Wellbeing Scale, and CARE Index mother–infant dyadic interaction synchrony. Results There were no significant differences in the change from baseline in Hospital Anxiety and Depression Scale and outwardly expressed irritability score between the combined active intervention groups and care-as-usual only [effect size (95% confidence interval) 0.03 (−0.24 to 0.29)], or between either Enhanced Triple P for Baby and Care-as-usual only [0.04 (−0.24 to 0.32)] or Mellow Bumps and care-as-usual only [0.01 (−0.27 to 0.30)]. Mellow Bumps holds promise for cost-effectiveness due to its potential for cost-savings relating to routine healthcare resource use. Limitations The main limitation of THRIVE was poor attendance of the groups (under 42%) offered to our Enhanced Triple P for Baby and Mellow Bumps participants. However, subgroup analysis adjusted for level of attendance suggested the results would not have been any different with increased participation. Conclusions With the population THRIVE involved, at a stage of pregnancy when a lot of support is offered (with many parenting interventions being offered without rigorous evaluation), we do not have evidence to recommend the rollout of Enhanced Triple P for Baby or Mellow Bumps. Future work Future work could explore the timing and intensity of interventions, intervention group composition, co-produced programmes, online sessions, and impact of addressing inequalities. Trial registration This trial is registered as Current Controlled Trials ISRCTN21656568 (www.isrctn.com/ISRCTN21656568). Funding This award was funded by the National Institute for Health and Care Research (NIHR) Public Health Research programme (NIHR award ref. 11/3002/01) and is published in full in Public Health Research; Vol. 13, No. 4. See the NIHR Funding and Awards website for further award information. Plain language summary What was the question? Women experiencing high levels of adversity, maternal depression and/or anxiety during pregnancy are more likely to feel stressed. That stress may lead to their baby being less able to cope with stress and may hamper the sensitive mother–baby relationship from developing. The THRIVE study aimed to explore ways that may help mothers protect their mental health and relationship with their babies. What did we do? Women were invited to take part in the THRIVE study by midwives and researchers. THRIVE took place in National Health Service Greater Glasgow and Clyde and National Health Service Ayrshire and Arran. The women who agreed to take part were randomly selected into 1 of 3 groupings. Group 1 received Enhanced Triple P for Baby plus care-as-usual, Group 2 received Mellow Bumps plus care-as-usual and Group 3 continued with care-as-usual only. We had two main aims. The first was to see if mothers’ mental health (anxiety and depression and reduce outwardly directed irritability) was better if they received either Enhanced Triple P for Baby or Mellow Bumps (interventions) compared with care-as-usual only. The second was to explore if the observed quality of relationship between mother and baby was stronger if they received an intervention versus care-as-usual only. What did we find? Our results showed that neither intervention was successful in improving either mental health or the observed quality of mother–baby interactions. Mellow Bumps is most likely to be deemed cost-effective due to its potential for cost-savings arising in routine healthcare resource use. What does this mean? We cannot recommend Enhanced Triple P for Baby or Mellow Bumps as being better than care-as-usual in the UK. We recommend that future work could explore the timing and intensity of interventions, intervention group composition, online sessions, co-produced programmes, and the impact of addressing inequalities. Scientific summary Background Evidence is growing that depression, stress, and anxiety in pregnant women can create adverse modifications to the fetus in utero that can permanently negatively influence the baby’s response to stress and disrupt the mother’s ability to be sensitive to her baby. As poor mother–baby interaction and maternal mental ill health often lead to child maltreatment, most postnatal interventions may not be able to undo some of the damage sustained by babies due to their mother’s and/or parents’ response to adverse circumstances during pregnancy. Women additional social and care needs in pregnancy (ASCNs e.g. mental ill health, domestic abuse, care experience) are likely to be more anxious, depressed and produce higher levels of stress-related hormones than women in more favourable circumstances. The Trial for Healthy Relationship Initiatives in the Very Early Years (THRIVE) study evaluated the effects of two parenting interventions on maternal mental health and mother–infant interactions for women who are identified as having ASCNs. Each intervention was provided in addition to care-as-usual (CAU), Enhanced Triple P for Baby (ETPB) and Mellow Bumps (MB) versus CAU. Objectives The overall aim of THRIVE is to evaluate the impact of the ETPB + CAU (ETPB) and MB + CAU (MB) parenting interventions on maternal mental health and mother–child relationships, when compared with CAU alone. In the longer term, THRIVE will also assess whether children whose mothers received ETPB or MB show reduced incidence of child maltreatment. Due to substantial amendments, our secondary research questions changed from protocol, mostly related to us not collecting data when the babies reached 18 months. The primary research questions that are addressed in this report were: Do mothers receiving ETPB or MB show significantly: Lower anxiety, depression and outwardly directed irritability compared with those receiving CAU only when their babies are about 6 months old? Higher maternal sensitivity scores in mother–infant interaction compared with those receiving CAU when their babies are about 6 months old? Secondary research questions included: Do infants whose parents receive ETPB or MB show more co-operative behaviour signs than those whose parents received CAU? Are either ETPB or MB more effective for different subgroups of parents? Are either ETPB or MB cost-effective for the NHS or society more broadly, in the long term? Do differences in programme fidelity, practitioners’ characteristics and motivation, mothers’ engagement, intervention mechanisms, and contextual factors affect mother and infant outcomes? Methods Study design A three-arm randomised controlled trial (RCT) comparing two parenting interventions, ETPB and MB, with CAU only. The study included women who received their care in NHS Greater Glasgow and Clyde (GGC) or NHS Ayrshire and Arran (A&A). Participants Women identified as having ASCNs during pregnancy based on the NHS GGC’s Special Needs in Pregnancy criteria were recruited to THRIVE. Subsequently, their babies and partner/accompanying person of choice were also invited to be involved in THRIVE. Interventions Both interventions aim to reduce maltreatment and improve sociodevelopmental outcomes for children, with content including information on infant development and incorporating cognitive–behavioural therapy to improve mothers’ coping in a general and parenting context. The set-up of the two interventions is also similar in terms of working with women from 20 weeks of pregnancy in groups of about eight participants with sessions delivered in local community venues by two trained practitioners. However, they have fundamental differences in focus and mechanism that may influence how effective they are at reducing child maltreatment. ETPB incorporates social learning principles, with an emphasis on families. ETPB includes fathers or an accompanying person of the participants’ choice and includes skills-based content focused on expectations and coping to meet the new challenges of parenthood while maintaining a happy family. MB is underpinned by attachment theory and designed to target mothers who are vulnerable in pregnancy. The focus of MB is on mothers, although fathers can be invited to one session, and the content focuses on nurturing mothers’ self-care, providing guided reflection, encouraging nurturing of the fetus/baby, engagement with the fetus/baby and synchrony in the mother–fetus/infant relationship. Primary study outcomes The following primary study end points were assessed between 6 and 12 months post partum using validated scales: Maternal mental health measured using the Hospital Anxiety and Depression Scale (HADS) and outwardly expressed irritability (I) measures from the Adult Wellbeing Scale (AWS). The dyadic synchrony between the dyad within the mother–baby interaction was measured using the CARE Index. The CARE Index uses video recordings of 3–5 minutes of mother–baby play interaction, which were blind coded by accredited coders. Secondary study outcomes The following secondary outcomes were assessed between 6 and 12 months postnatal: HADS depression subscale. HADS anxiety subscale. AWS outwardly expressed irritability subscale. CARE Index co-operative behaviour. The intervention group was entered in the regression models in the following ways: Binary as active intervention (ETPB or MB) versus CAU. Binary as ETPB versus MB (on subgroup receiving any active intervention). Categorical with the three categories ETPB, MB and CAU. Economic evaluation outcomes EuroQol-5 Dimensions (EQ-5D) total domain score and visual analogue scale (VAS). Costs were adjusted using generalised linear models with gamma family and log link to account for covariates and skew distribution of cost. Then incremental cost-effectiveness ratio (ICER) in the cost–utility analysis was calculated as the ratio of incremental cost difference and incremental quality-adjusted life-years (QALYs) difference (ETPB vs. CAU and MB vs. CAU). Process evaluation A realist process evaluation (PE) was undertaken and utilised a range of methods, including participant observation, interviews and questionnaires to examine ‘what works for whom in what circumstances?’ The main themes were independently coded across a sample of interviews by two researchers, Rosaleen O’Brien (RO) and Katie Buston (KB), using NVivo 10 (QSR International, Warrington, UK). A coding framework was then agreed by RO and KB and discussed/refined with the THRIVE PE subgroup, based on the key questions for the PE, developed prior to implementation, with the addition of novel areas identified through data collection. Recruitment data collection and analysis Participants were recruited to THRIVE over 53 months from January 2014 to May 2018. Three main recruitment strategies were used: referral from health or social care practitioner or third-sector organisation, who screened for eligibility and interest for participation in the trial from October 2015 (following a protocol amendment), participants could be recruited in person in a clinic or community setting by a member of the THRIVE team; and self-referral to the trial by responding to poster advertisements in community/healthcare setting, targeted web-based advertisement, or social media advertising. For strategies 2 and 3, potential participants were screened for eligibility and a referral form was completed by THRIVE staff. Following receipt of a referral form, or a note of, details of potential participants were entered in the study database and the NHS GGC Clinical Research Facility were contacted to determine whether the pregnancy was continuing. Once confirmed, a member of the THRIVE team contacted potential participants to arrange an appointment at which they were afforded the opportunity to ask questions about the research, and on agreeing to participate, were consented to trial and had baseline measures completed. Results From the 964 referrals received, 488 participants were recruited to THRIVE (marginally below the aspired 500) and completed baseline assessments. The intention-to-treat (ITT) population included 450 participants. Among the ITT population, 135 participants who received ETPB, 131 who received MB and 53 who received CAU only, completed follow-up measures at 6–12 months post partum. We therefore reached statistical power for our primary and secondary analyses. The demographic and baseline characteristics of THRIVE participants were balanced across groups. The mean age was 26.7 years, 63.0% of participants were in the Scottish Index of Multiple Deprivation quintile 1 (i.e. the 20% most deprived areas) and 94.4% reported white ethnicity (the 2011 Scottish Census showed that 96% reported a white ethnicity). This was the first reported pregnancy for 45.1% of participants, and 40.2% of all participants reported that their pregnancy was planned by both parents. Primary study outcomes Hospital Anxiety and Depression Scale plus irritability At baseline, the mean standard deviation (SD) HADS + I score was 0.31 (0.17) for ETPB, 0.31 (0.17) for MB and 0.32 (0.17) for CAU only. Mean changes from baseline to follow-up (6–12 months post partum) were −0.03 (0.18) for ETPB, −0.03 (0.16) for MB and −0.04 (0.15) for CAU only. These mean scores suggest that on average the participants were not depressed, anxious, or irritable at the start of the trial or at 6–12 months post partum. There were no significant differences in the change from baseline in the HADS + I score between the combined active intervention groups and CAU only {effect size [95% confidence interval (CI)] 0.03 [−0.24 to 0.29]}, or between either ETPB and CAU only [0.04 (−0.24 to 0.32)] or MB and CAU only [0.01 (−0.27 to 0.30)]. The same pattern of results was found in the per-protocol (PP) analysis and in analyses conducted to assess the sensitivity of HADS + I to missing values and clustering within intervention groups. Mother–infant dyadic interaction synchrony The mother–infant dyadic interaction synchrony was assessed at follow-up when babies were 6–12 months old using the CARE Index dyadic synchrony score. Mean (SD) dyadic synchrony scores were 6.6 (2.7) for ETPB, 6.4 (2.5) for MB and 7.1 (2.6) for CAU. These scores suggest the mother–infant scores in THRIVE were slightly below average in terms of maternal sensitivity and baby’s co-operativeness. There were no significant differences in the CARE Index dyadic synchrony score between the combined active intervention groups and CAU only [effect size (95% CI) −0.20 (−0.52 to 0.12)], or between either ETPB and CAU only [−0.17 (−0.52 to 0.18)] or MB and CAU only [−0.24 (−0.59 to 0.11)]. The same pattern of results was found in the PP analysis, analyses conducted to assess the sensitivity of HADS + I to missing values and clustering within intervention groups and after removing dyadic synchrony scores generated from episodes of caregiving rather than play. Neither mediator and moderator nor subgroup analysis was able to show any significant results. Secondary study outcomes When the HADS + I score was broken down into its subscales, there were no significant differences between treatment groups for HADS depression score, HADS anxiety score, or outwardly expressed irritability score. There were also no significant differences between groups for other CARE Index outcomes, including the mother’s sensitivity score or the co-operative behaviour score. Several other secondary outcomes were assessed, including support and control in childbirth, post-partum partner support, parental regulation, emotional regulation, maternal attitudes to marital relationships, perceived social support, parenting self-confidence, maternal attitudes, parental coping, response to infant crying and post-partum bonding. The only secondary outcome for which significant differences were observed between treatment groups was the Cognitive Emotional Regulation Questionnaire. Significantly lower scores were observed for ETPB versus CAU only for the focus on thought subscale (better) and the focus on planning subscale (worse) and for MB versus CAU only for the acceptance subscale (worse), the focus on thought subscale (better) and the positive reappraisal subscale (worse). Economic evaluation outcomes Quality-adjusted life-years Participants had lower utility at follow-up one for both ETPB and MB groups compared to the CAU arm; the adjusted mean utilities at ETPB group and MB group were 0.011 (95% −0.079 to 0.063) and 0.021 (95% −0.095 to 0.061) lower than CAU. There was no statistically significant difference between QALYs gained in each group over the 1-year period. Cost–utility results Compared to the CAU, both ETPB and MB group had a decreased point estimate of cost, at −£703 (−£3725, £2045) and −£1570 (−£4487, £703), respectively, which demonstrated a potential for cost-saving. Please note, this is not cost-effective in the traditional sense, as there were no quality-of-life savings, but in terms of cost-savings and against a background of great uncertainty in the effectiveness outcome. When hospitalisation costs were excluded in the sensitivity analysis, MB was no longer a cost-saving and CAU became the most cost-effective option due to its lower cost. Both ETPB and MB groups did not show any QALY gain compared to the CAU group; both with a slight, statistically insignificant, QALY loss of 0.23 over the 1-year period. Given both the mean incremental cost and mean QALY gain were negative, net monetary benefit was used instead of the ICER to determine the cost-effectiveness against the threshold £20,000. It shows that MB had the highest point estimate of the net monetary benefit, at £6035 (95% CI £4120 to £7800), followed by ETPB at £5167 (95% CI £2551 to £6965) and CAU at £4905 (95% CI £1960 to £7051). At £20,000 per QALY threshold, the probability for MB to be cost-effective was highest, at 65%, and 22% for ETPB, and 13% for CAU. Process evaluation The main conclusion, anticipated in advance, was that there would be limited evidence of effectiveness revealed by the outcome evaluation. ETPB appeared to be easier to implement as theorised, at least during the antenatal phase, but fidelity was by no means perfect, and attendance was concerning for both interventions. Limitations The main limitation of THRIVE was poor attendance at the groups (under 42%) offered to the ETPB and MB participants. However, a subgroup analysis that adjusted for level of attendance did not suggest that the results would have been different if participants had taken up more of the interventions. Conclusions THRIVE was not able to demonstrate an impact of ETPB or MB on prespecified primary outcomes among participants with additional social and care needs, compared with CAU. The economic evaluation suggests that MB may help reduce routine health care, but this finding needs to be interpreted with caution. There is clearly much need for support among the THRIVE population, and CAU works intensively with families at this stage of life, it may be that ETPB or MB was not able to add much more value to the CAU being offered. We suggest working with those with lived experience of additional social and care needs to explore other avenues of support that can be complementary to CAU. Our recommendations for future action and research, in numbered order, are to explore: Timing and intensity More research is needed to identify the best timing for early years parenting interventions, to complement CAU, and ideally programmes that can adapt to the intensity required for parents with different levels of additional social and care needs. Group composition It would be interesting to explore the benefit of more homogeneous groups’ composition rather than the artefactually heterogeneous groups that were evaluated in THRIVE. Complementary app/online sessions Some women could not attend group sessions due to work, child care or ill-health issues, but may have engaged with online information. So, exploring hybrid engagement opportunities may be useful in the future. Women with mental ill health or social anxieties may also prefer online options. Co-produced programmes Enhanced Triple P for Baby and MB were developed by experts in clinical psychology for participants with additional care and support needs but they may not be experts in the THRIVE population. A higher level of involvement with people with lived experience in the development of interventions may help programmes to be delivered when most beneficial to potential participants. It is likely that there is no one size fits all and that support for those with additional social and care needs may need to be able to respond flexibly to bespoke and complex life situations. Exploring ways to support communities to help themselves may be beneficial. A literature review of co-production in mental health suggests that this is a promising approach. Addressing inequalities There is a large body of evidence showing the impact of inequalities on health. Moving towards a more egalitarian society may lead to less pressure on the NHS and need for programmes, such as ETPB and MB. Trial registration This trial is registered as Current Controlled Trials ISRCTN21656568. Funding This award was funded by the National Institute for Health and Care Research (NIHR) Public Health Research programme (NIHR award ref. 11/3002/01) and is published in full in Public Health Research; Vol. 13, No. 4. See the NIHR Funding and Awards website for further award information.https://doi.org/10.3310/KYMT5407mental healthmother-infant interactionparentingpregnancyrandomised controlled trialvulnerable women |
| spellingShingle | Marion Henderson Anja Wittkowski Katie Buston Karen Crawford Alice MacLachlan Alex McConnachie Emma McIntosh Claudia-Martina Messow Catherine Nixon Rosaleen O’Brien Shona Shinwell Daniel Wight Yiqiao Xin Rachel Calam Ruth Dundas James Law Helen Minnis Lucy Thompson Philip Wilson 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 Public Health Research mental health mother-infant interaction parenting pregnancy randomised controlled trial vulnerable women |
| title | 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 |
| title_full | 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 |
| title_fullStr | 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 |
| title_full_unstemmed | 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 |
| title_short | 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 |
| title_sort | 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 |
| topic | mental health mother-infant interaction parenting pregnancy randomised controlled trial vulnerable women |
| url | https://doi.org/10.3310/KYMT5407 |
| work_keys_str_mv | AT marionhenderson evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT anjawittkowski evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT katiebuston evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT karencrawford evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT alicemaclachlan evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT alexmcconnachie evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT emmamcintosh evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT claudiamartinamessow evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT catherinenixon evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT rosaleenobrien evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT shonashinwell evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT danielwight evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT yiqiaoxin evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT rachelcalam evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT ruthdundas evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT jameslaw evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT helenminnis evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT lucythompson evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents AT philipwilson evaluationofparentinginterventionsforthosewithadditionalhealthandsocialcareneedsduringpregnancythriveamultiarmrctwithembeddedeconomicandprocesscomponents |