Randomized controlled trial and economic evaluation of nurse-led group support for young mothers during pregnancy and the first year postpartum versus usual care

The study’s objective was to determine whether gFNP, compared to routine antenatal and postnatal care, could improve parenting, and hence reduce risk factors for maltreatment in a vulnerable population and also be cost-effective.

The design was a multi-site, blinded, randomized controlled, parallel-arm trial and prospective economic evaluation with eligible pregnant women at seven locations in England allocated (minimized by site and maternal age group) to one of two arms: (1) gFNP and (2) usual care. All participants (both arms) were eligible to receive usually provided publicly funded health and social care. Baseline data were collected in early pregnancy with follow-up when infants were 2 and 6 months old and the primary endpoint at 12 months.

The study was granted approval by the UK National Research Ethics Service Committee South West-Frenchay (REC reference 13/SW/0086).

Family Nurse Partnership teams were eligible to take part in the trial if they were experienced in delivering FNP and had at least one FN who had notified their intention to practice as a midwife. Sites responded to an invitation from the FNP National Unit to take part, demonstrating that sufficient women of the relevant age and parity had given birth in the previous year, and confirming good links with community midwifery, who also signed the expression of interest. Seven sites agreed to take part, representing London (two sites), the Midlands (two sites), the North East (one site) and the North West of England (two sites).

Women eligible for the trial were expectant mothers with a gestation of 16 to 20 weeks, with expected delivery dates (EDDs) within approximately 10 weeks of each other, for each group in each site. They were either aged below 20 years at their last menstrual period (LMP) with one or more previous live births, or aged 20–24 at LMP with no previous live births and with low educational qualifications, defined as neither mathematics nor English language GCSE at grade C or higher; if both, then no more than four GCSEs at grade C or higher. They had to be able to provide consent and to speak English. Women who had previously received FNP and those with psychotic mental illness were not eligible.

Full details of the recruitment can be found in the study protocol [15]. The trial commenced in February 2013, recruitment and baseline data collection commenced in July 2013, continuing to September 2014. Recruitment was in three phases, each separated by approximately 12 weeks, so that FNP teams could offer more than one group. Potential participants were identified from records by community midwives, and then after some difficulties in identifying a sufficient number [16] also by FNP midwives and Clinical Local Research Network (CLRN) research midwives, based on their age, parity and gestation from midwifery records. After written or telephone agreement for the researcher to contact the potential participant, a telephone contact confirmed eligibility by asking about educational qualifications and any psychotic mental illness. If they were eligible and interested in the study the researcher arranged a home visit so that written consent could be obtained and baseline information collected. Following baseline data collection their contact details were entered into a secure database which generated an identification number.

Randomization was overseen by the London School of Hygiene and Tropical Medicine Clinical Trials Unit (LSHTM CTU). The unique identification number (which included a site identifier) and age at LMP of eligible consenting mothers-to-be were passed to the central randomization service at the Health Service Research Unit (HSRU), University of Aberdeen using an automated telephone procedure. Minimization criteria (site, and age group below 20 years, 20–24 years) were used to ensure a balance of key prognostic factors. Due to low recruitment, randomization was changed from 1:1 to 2:1 in December 2013 after approval from the Trial Steering Committee (TSC), Data Monitoring Committee (DMC) and Research Ethics Committee (REC). Allocation to one of two arms was securely computer-generated and delivered by email to LSHTM which conveyed the information to study participants by post and conveyed to each gFNP team the names and contact details of women allocated to the intervention arm by fax or password-protected email, receiving confirmation of receipt by email. Participants could not be blind to allocation as they knew whether or not they had been offered gFNP, and gFNP practitioners were not blind to the intervention participants but had no knowledge of the control group. Service providers other than gFNP personnel may or may not have been blinded depending on what study participants chose to discuss when receiving usual care. The research team members responsible for collecting the questionnaire data, including the principal investigator (PI), the trial manager, the interviewers and also those scoring the videos for the CARE Index, were blind to treatment allocation.

Study intervention: Group FNP (gFNP) is designed to run from the first trimester of pregnancy until infants are 12 months old with 44 group meetings in the curriculum, 14 covering pregnancy and 30 covering infancy [17]. It is delivered to a group of women living in relatively close proximity to each other, with similar expected delivery dates (range 8–10 weeks) [9]. Meetings last around 2 h, are held in children’s centers, health centers or other suitable community facilities in the local areas served by the FNP teams. Sessions were facilitated by two experienced FNP FNs one of whom had notified their intention to practice as a midwife. The two FNs exchanged the roles of active leader (facilitating a topic and activity) and active observer, noticing behaviors and body language of members and stepping in to support the leader and maintain a positive and inclusive group environment.

The gFNP program includes content to: improve maternal health and pregnancy outcomes, improve child health and development by helping parents provide more sensitive and competent care; and to improve parental life course by helping parents develop effective support networks, plan future pregnancies, complete their education, and find employment [9]. The curriculum domains are: mother’s personal health; the maternal role; maternal life course: family and friends; environmental health; and related health and human services, with referrals made when necessary. In general all the content domains are covered to a greater or lesser extent in each of the meetings, varying according to a detailed manual. The gFNP curriculum materials and activities were modified from those used to deliver FNP to reflect group administration. They were designed to avoid a lecture context but to facilitate interaction between group members and between group members and the nurses, providing a range of engaging, often “hands-on” activities. In particular, gFNP had a particular focus on enhancing social support and social networks through dialog between group members, which is not a specific focus of home-based FNP [9, 17].

Specific to the gFNP program and following national guidelines [18], the FN midwife provided routine antenatal care during the meeting, taking an approach based on the Centering Pregnancy program [12, 19, 20] which encourages women to monitor their own health (e.g., by testing their own urine, listening to the fetal heartbeat). The Centering Pregnancy approach was perceived to correspond well with the gFNP aims in that both focus on developing self-efficacy and encouraging women to be more self-aware [9]. Once infants were born both FNs were involved in routine infant checks, conducted according to the UK National Health Service (NHS) Healthy Child Program (HCP) [10].

Appreciation of the diversity of group members is central to how the nurses deliver the content, especially for some emotive topics such as “safe relationships for our children” [9]. While there is a curriculum for each meeting the nurses were sensitive to the need for “agenda matching” related to particular issues raised; this requires the practitioners to listen to the issues that are uppermost for the group members and agree how these can be met while at the same time ensuring that the session agenda is realized and behavior adaptation is progressed for everyone. In addition to modeling of infant care, they model respectful relationships and turn-taking [21], behaviors that are expected to be of benefit to any group members with poor social skills, especially if they are experiencing difficult interpersonal relationships [9]. Study participants allocated to gFNP could also access any aspect of the HCP usual care that they wished, independently or with the guidance of the gFNP nurses.

The aims of the HCP are to develop strong parent-child attachment and positive parenting, resulting in better social and emotional wellbeing among children; care that helps to keep children healthy and safe; healthy eating and increased activity, leading to a reduction in obesity; prevention of some serious and communicable diseases; increased rates of initiation and continuation of breastfeeding; readiness for school and improved learning; early recognition of growth disorders and risk factors for obesity; early detection of – and action to address – developmental delay, abnormalities and ill-health, and concerns about safety; identification of factors that could influence health and wellbeing in families; and better short- and long-term outcomes for children who are at risk of social exclusion.

There is a focus on supporting mothers and fathers to provide sensitive and attuned parenting, in particular during the first months and years of life. From the 12th week of pregnancy women are encouraged to see a midwife or maternity healthcare professional for a health and social care assessment of their needs, risks and choices.

Data collection was conducted by field researchers making four visits to participants’ homes (baseline in early pregnancy, when infants were 2 months, 6 months and 12 months old), when they administered structured questionnaires and at 12 months (if additional consent was given) they made a 3- to 5-min video-recording of the mother and infant together, presented with a standardized set of toys. Participants were given a “High Street” voucher for £20 at each home visit. Measures used at each time point are presented in Table 1 with full details in the published protocol [15]. Researchers sent completed questionnaires by post to LSHTM CTU and checks were made for receipt. Videos of play interactions were transferred by fieldworkers from camera to encrypted USB flash drives with AES 256-bit military-level security, sent by recorded delivery to the PI, with files deleted from the camera by the fieldworkers. Recordings were decrypted and saved with full anonymization of filenames on a dedicated drive separate from any other study information.

Two primary outcome measures at 12 months were used: the revised Adult-Adolescent Parenting Inventory (AAPI-2) [22], a 40-item self-report measure of attitudes to parenting previously found to discriminate between abusive and non-abusive parents [22]; and maternal sensitivity rated by observation using the CARE Index, which can range from 0 to 14 with scores of 0–4 in the risk range for child maltreatment [23]. Eight pre-specified secondary outcomes assessed infant cooperativeness [23]; maternal depression [24]; stress [25]; sense of competence [26]; social support [27]; smoking, alcohol and drug use; relationship violence; and breastfeeding to 6 months (see [15] for full details and Table 1).

To assess maternal quality-adjusted life years (QALYs) for the economic evaluation, the EuroQol EQ-5D-5 L measure [28] was administered at each time point (see Table 1). This five-dimension health status classification system can be converted to a multi-attribute utility score by applying a national tariff [29].

Adverse events were assessed using information from research midwives, maternal interview or pre-paid change of circumstances cards, and categorized as adverse (hospitalization of mother or infant other than for delivery) or severe (loss/termination of the pregnancy, congenital anomaly or birth defect, persistent or significant disability and death of either mother or infant). Severe events were reported to the REC within 15 days of the PI becoming aware of the event.

For the individually randomized trial we proposed to recruit sufficient mothers and babies (families) to detect a difference between arms of 0 · 5 standard deviations (SDs), with 90% power at a significance level of 0 · 05 (two-tailed), considered to represent a moderate size of effect [30]. Basing calculation on the AAPI-2 [22], very conservatively assuming a correlation of 0 · 4 between pre- and post-intervention scores, at least 71 families were needed in each arm of the trial to detect this difference. Allowing for an expected 30% dropout rate (based on the first two applications of the program in England) we would need to recruit 84 families per arm of the trial. We therefore proposed, conservatively, to recruit a 100 families per arm (N = 200). The proposed sample size would similarly allow us to detect a change of approximately 0 · 5 SDs in the CARE Index maternal sensitivity score [23]. If this was achieved we expected to be able to detect a difference at follow-up between arms of the trial of approximately 1 · 2 with 90% power and a 5% level of significance. Based on additional calculations when changing the allocation ratio following recruitment difficulties [16], a sample size of 166 participants would have 81% power to detect an effect of 0 · 5 SDs in the primary outcomes.

Primary analyses were by intention-to-treat (ITT), i.e., based on random allocation regardless of receipt of program. Follow-up was not restricted to those receiving the program. Analyses included adjustment for baseline measure of the outcomes where available (ANCOVA) and also an “adjusted” analysis that additionally included adjustment for the minimization factors site and maternal age group. The complete case analysis provides an unbiased estimate of the ITT effect under a “missing-at-random” assumption but, in addition to the primary analyses, complier average causal effect (CACE) analyses [31] were conducted to estimate a measure of the effect of the intervention on those participants who received it as intended by the original allocation. Where outcomes were collected at multiple time points, random-effects models, using a likelihood-based approach, were simultaneously fitted to the outcomes at all the time points that they were measured so that data from all the participants contributed to the analyses, even if there were missing data at some of the follow-up points The statistical analysis plan as agreed with the DMC in December 2014 was for the primary analysis to exclude all data outside pre-specified windows, i.e., after 12 months plus 60 days, 6 months plus 45 days and 2 months plus 30 days. However, in addition, sensitivity analyses were conducted with all data including those outside the windows. For the primary outcome of the AAPI-2 [22] a linear regression model was used to estimate a mean difference in 12-month AAPI-2 scores between the two arms of the trial. For the primary outcome maternal sensitivity score [23] a mixed-effects model was planned with a random effect at the mother level (to allow for multiple births) to estimate a mean difference in maternal sensitivity score between the two arms of the trial. However, only one set of twins was available for this analysis and their responses were identical. Therefore, it was not possible to include a random effect and the analysis was carried out at the mother level using a linear regression model.

For the secondary outcomes, appropriate generalized linear models were used to examine the effect of the intervention. Odds ratios and mean differences are reported with 95% confidence intervals (CIs). As for the primary outcomes, where continuous measures were available at baseline they were adjusted for in the analysis with additional analyses adjusted for site and maternal age group. Where there was evidence of non-normality in the continuous outcome measures, non-parametric bootstrapping with 1000 samples was used to estimate the effect of the intervention and bias-corrected CIs are reported [32]. Where this was done, p values were estimated using permutation tests. Statistical analyses were conducted using Stata version 14. An independent Data Monitoring Committee (DMC) oversaw the trial.

A within-trial economic evaluation was conducted from a UK NHS and personal social services perspective. Primary research methods estimated the costs of delivering gFNP, including development and training of accredited providers, the cost of delivering the group sessions, participant monitoring activities, and any follow-up/management. This primarily involved asking each of the gFNP practitioners in each site to prospectively complete detailed weekly activity logs outlining the cost of delivering each gFNP session, including costs associated with preparation time, program delivery time, indirect administrative activities, home visits and telephone contacts, as well as gFNP-related training and supervision activities. The weekly activity logs also recorded the mode, distance and time spent traveling by each practitioner as a result of gFNP-related activities. They also recorded additional expenditures associated with refreshments, materials, cards or gifts, participant travel, partner travel, child-care costs and miscellaneous expenditures associated with weekly gFNP-related activities. The costs of venue hire were estimated separately within each site. The total costs of delivering the gFNP program across each group and site were subsequently converted into group and site-specific estimates of average cost per session per attending woman using separately collected attendance data for each group within each site. Broader resource utilization was captured through maternal questionnaires completed at each follow-up point and provided profiles of hospital and community health and social services received by each trial participant (and, in the case of the postpartum questionnaires, their baby). Information was also collected regarding use of legal services and costs borne by the trial participants or their family members or friends as a result of the trial participants’ (and, in the case of the postpartum questionnaires, their baby’s) health status, over the relevant time horizons. Unit costs (£, 2014–2015 prices) were collected from national sources in accordance with guidelines and attached to resource use. Health utilities generated from EuroQol EQ-5D-5 L [28] responses were used to estimate QALY profiles for each woman, calculated as area under the baseline-adjusted utility curve, assuming linear interpolation between utility measurements. Costs and QALYs accrued by each trial participant beyond the first 12 months of follow-up were discounted at 3 · 5% as recommended by the National Institute for Health and Care Excellence (NICE) [33]. Multiple imputation using chained equations [34] and predicted mean matching was carried out on the EQ-5D-5 L, as well as cost estimates, at 2, 6 and 12 months postpartum [35]. Mean (standard error (SE)) costs by cost category and mean (SE) total costs were estimated by trial allocation arm for all time periods.

Cost-effectiveness results are reported as incremental cost-effectiveness ratios (ICERs), calculated as the difference in mean costs divided by the difference in mean outcomes (QALYs or maltreatment outcome measures) between the trial comparators. The nonparametric bootstrapping approach was used to determine the level of sampling uncertainty surrounding the mean ICERs by generating 10,000 estimates of incremental costs and benefits, represented graphically on four quadrant cost-effectiveness planes. Cost-effectiveness acceptability curves (CEACs) showing the probability that the gFNP program is cost-effective relative to standard care across a range of cost-effectiveness thresholds were also generated based on the proportion of bootstrap replicates with positive incremental net benefits. Several sensitivity analyses were undertaken to assess the impact of uncertainty surrounding components of the economic evaluation. These included: (1) adopting a wider societal perspective, (2) restricting the analyses to complete cases, and (3) recalculating the average cost per gFNP session per attending woman by varying (1) the mean number of gFNP sessions attended to the highest and lowest observed mean values across all groups across all sites and (2) the number of gFNP group participants to the highest and lowest number of observed values across all groups across all sites. Subgroup analyses were conducted for the main cost-effectiveness results by: (1) program completers (no, yes) where women who participated in a pre-specified number of group sessions (at least 17 sessions) were regarded as “program completers” and (2) program phase (1, 2, 3) to test whether organizational learning may have influenced the cost-effectiveness of the gFNP program. Analyses for the economic evaluation were conducted using Stata software version 14.