Discussion

Summary of findings

The phase 1 study found that nurseries reported a generally good standard of nutrition, but the extent to which nutrition guidance was used varied between settings, with some using no guidance. Most nurseries had space for children to be physically active, which was usually outdoors. There was variation in the extent to which the outdoor space was used regardless of weather, or only on dry days. Screen time was limited across settings, with few including television time in their daily routines. Children’s oral health was highlighted as a growing concern in nurseries, and this concern was reinforced by an NHS oral health specialist who was a member of our group of expert advisors to NAP SACC UK. Adaptations to NAP SACC were made in the light of the phase 1 study and NAP SACC UK at Home was created.

In the phase 2 feasibility RCT, recruitment of nurseries (31.6%) and children (35.3%) was lower than the progression criteria of 40%, but both were exceeded in one of the two study areas. No nurseries withdrew from the study. The intervention was delivered as planned in five out of the six intervention nurseries with high levels of feasibility and acceptability (there were particular circumstances in the sixth nursery), with the exception of the home component. It was feasible to recruit and train health visitors to deliver the intervention, but health visitors reported they may not have capacity to deliver the intervention alongside their usual workload.

The trial methods and design were acceptable and feasible. The number of children lost to follow-up was 24 (14.2%), because of withdrawing consent (1.8%), child refusal to participate on the day (1.2%), children moving to primary school (6.5%) or children moving nursery (4.7%). Eighty-five per cent (n = 145) of children were followed up for data collection, but not all children provided all data at baseline and follow-up. Although the study was not powered to measure effectiveness, accelerometer-measured total PA and MVPA were higher in the intervention arm than in the control arm after the intervention. The ICC for total activity was 0.02 (95% CI < 0.0001 to 0.95). The average cost of delivering the intervention was estimated at £1184 per nursery; nurseries incurred on average an additional £717 for staff attendance at two workshops.

The two substudies found the mediator questionnaire to be reliable and the methods of collecting diet data using photography to estimate portion size to be feasible.

Progression criteria

Chapter 5, Table 27, outlines the trial outcomes for each of the progression criteria. These criteria were met in full except for the recruitment of nurseries and children and the feasibility of the NAP SACC UK at Home component. Although the overall recruitment rates for nurseries and children were below the target 40%, both were exceeded in North Somerset, where the length of recruitment was longer, and the independent TSC agreed that these data showed recruitment is feasible.

Intervention acceptability

The intervention was implemented with high fidelity except that (1) one nursery did not implement the intervention because of staff workload (we concluded that the issues for this nursery were not generalisable and would add visits to all nursery managers and staff at recruitment to avoid this in a future trial) and (2) the digital home component was used by only 14% of parents. The process evaluation showed that the NAP SACC UK intervention was feasible to implement in a UK nursery. The mean number of staff per nursery attending workshops was seven. The workshops and contact with health visitors were highly valued. The mean number of goals set was eight. Reported changes included menu modifications, reducing portion sizes and sugary snacks, role modelling PA and eating, and active story telling. The intervention was highly acceptable to the NAP SACC UK partners and they reported that they enjoyed delivering the intervention. Concerns were raised by NAP SACC UK partners as to how, if rolled out, this would fit into their (health visitor) current workload as they are already overstretched. Suggestions included having dedicated NAP SACC UK health visitors (separate to normal health visitors duties), using other staff groups (e.g. health improvement staff) or having a generic job description for a set of key skills.

The intervention was found to be highly acceptable to nursery managers and staff. Workshops were frequently mentioned as the part of the intervention most valued by these groups, along with the contact with a named health visitor. Parents who were interviewed were supportive of the research process taking place in the nursery, but were usually unaware of the changes that had taken place because of the intervention.

Research feasibility and acceptability

The trial design and methods were highly acceptable. Some nursery staff did not fully understand the randomisation process, but the majority of staff found it an acceptable process. Staff reported that the data collection process was acceptable and caused little disruption; similarly, parents reported the data collection to be acceptable. No nurseries left the trial, and only 14.2% (24/169) of children were lost to follow-up, primarily because they left the nursery (either to a different nursery or to school).

Indication of affordability

The study included piloting cost-effectiveness measures through the use of logs and questionnaires to estimate intervention costs. The mean cost of the intervention was £1184.39 (SD £209.48) per nursery or £27 per 2- to 4-year-old child. The majority of this cost was attributed to the staff time required to deliver the intervention. Depending on who delivers the workshop training and the support to nurseries, local authorities could potentially absorb most of the costs of delivering the intervention within currently commissioned services (e.g. health visiting) or provision by local authority staff. Alternatively, the intervention could be funded by public health or early years local authority departments and commissioned as an externally provided service.

Suggestion of promise

The accelerometer data showed suggestion of promise for the intervention increasing MVPA and total activity on nursery days in the intervention arm compared with the control arm, which was not seen to the same extent on non-nursery days. We would aim to increase the number of children with valid wear time of the accelerometers by using methods adopted in a study by Pate et al.48 in nurseries in the USA. In Pate et al.’s study,48 research staff returned to the nursery each morning to give replacement accelerometers to children who did not arrive wearing the accelerometer given the previous day. Furthermore, families were given a $25 gift card at the end of each data collection period.

There was further suggestion of promise for the knowledge and motivation of parents and nursery staff, reduction in screen time and sedentary activities at home, increase in the quality of snack provision, and decrease in the proportion of children who are overweight or obese.

Implications for future research

Phase 3 trial

The NAP SACC UK intervention and trial methods were feasible and acceptable and we found suggestion of promise in the feasibility trial. Therefore, only minimal refinements to intervention content are required. The feasibility trial supports the case for a phase 3 effectiveness trial. The feasibility work has informed the design of a full-scale trial and a multicentre trial has been funded by NIHR (PHR NIHR 127551) which will start in July 2019, informed by a range of sample size calculations (see Appendix 11, Table 47) to detect a 17-minute increase in total activity on nursery days immediately post intervention.

The PICO (Participants, Intervention, Control, Outcomes) for a full trial is shown in Box 1, and Table 29 outlines the key steps of the NAP SACC UK interventions as they will be used in the full trial over 12 months. Steps 1–4 will take place over 6 months and steps 5–7 over the following 6 months.

Strengths and limitations

This is a study in the UK that has tested the feasibility of a combined nutrition and PA environmental intervention in nurseries, with the additional component of oral health. The adaptation of the intervention from the USA was well informed by communication with parents, nursery managers and providers of services in early years settings.

For a feasibility study, it was a strength that two local authorities were involved, which included a range of city, small town and rural areas and neither area was adjacent to the university (where settings are more likely to be exposed to opportunities to take part in research), therefore giving a good indication of recruitment in other areas of the country. The intervention and trial methods were tested in a good number of nurseries and with a large number of children. Rates of nursery and child recruitment were reasonable. The retention of nurseries was excellent and retention of children was good where the children stayed within the nurseries; we deliberately tested the feasibility of following children beyond their time in nursery with the progression to school.

Objective measures of PA, diet, zBMI and the nursery environment were collected from participants at two time points. The study incorporated a detailed process evaluation that has allowed us to address the progression criteria from the perspectives of parents, nursery staff and NAP SACC UK partners. The study further benefited from a detailed test of measures of HRQoL and cost to inform a future assessment of cost-effectiveness.

The intervention was provided free to nurseries. Although the nurseries incurred the cost of staff time to take part, none of the six intervention nurseries raised this as a concern and only one nursery decided not to take part prior to recruitment because of this particular concern. All children in nurseries were potentially able to benefit from improvements made by the nurseries.

We have reported the intervention in accordance with the TIDieR framework and CONSORT reporting guidelines (with the CONSORT flow diagram; see Figure 4).

The data collection using CADET allowed us to assess food consumption by food group but not by portion size, because some nurseries reported making changes in portion size and this was not measured objectively. The number of children with complete accelerometer data at baseline and follow-up for the required wear time was low. Response rates from parents to complete the parent questionnaire were lower at follow-up. Both could be addressed through use of incentives and different protocols for data collection in a full trial;48 we would not intend to measure parental costs of diet and PA, health-care use or time off work (because there were too many missing data, these data were highly variable and these costs were unlikely to be affected much by the intervention).

We were not aware that there were two other quality improvement programmes available to nurseries in Gloucestershire prior to recruitment and, therefore, these were not part of the exclusion criteria for recruiting nurseries. Many nurseries in Gloucestershire were taking part in the ‘Bristol Standard’ and the pilot ‘Smiles Better award’. The Bristol Standard addresses 10 dimensions of educational quality (including health dimensions) across the nursery environment through a process of self-reflection, evaluation and goal-setting by nursery staff. Two out of the six recruited nurseries in Gloucestershire were also involved with the Bristol Standard. The Smiles Better award addresses oral health. The recruitment rate may have been higher had we excluded nurseries participating in other non-health quality improvement programmes in Gloucestershire. This was discussed after recruitment with the TSC; the TSC advised it would have been reasonable for nursery participation in quality improvement programmes, other than NAP SACC UK, to have been included in the exclusion criteria. If this had been an exclusion criterion the nursery recruitment rate would probably have been higher. The eligibility criteria for a future trial will need to consider whether or not nurseries should be excluded if they are already committed to other quality improvement programmes because of the capacity of nurseries to take part in multiple programmes.

We did not recruit any large nurseries in the highest deprivation group. Nurseries did not need to give a reason for not taking part and, therefore, we have limited information as to why these nurseries chose not take part. It is possible that the large nurseries in more deprived areas have more children who do not have English as a first language and have more children aged 2 years accessing the free child care that is means-tested for 2-year-olds. Therefore, these nurseries may have a greater workload and taking part in a research study may be regarded to be an additional burden. For a full trial, as part of the recruitment, we will include large and deprived nurseries in a recruitment video and ensure that issues regarding the benefits of reducing health inequalities are highlighted, to help large nurseries in deprived areas to identify with other nurseries that took part and to make the study feel manageable, attractive and of importance to their children.

In this study, zBMI was used as the measure of adiposity for children. We recognise that this cannot provide data on fat mass and fat-free mass, which may be important and could be detected by other approaches, such as dual-energy X-ray absorptiometry (DEXA) scans.115 However, these methods are not applicable for use within a large field study and it would not be feasible to use such an approach in a definitive trial without substantial increases in participant burden and research cost. Thus, despite the potential limitations zBMI is the most appropriate measure for this study.

Conclusions and recommendations

The NAP SACC UK intervention (without the home component) and trial design have been found to be acceptable to nursery staff, parents and NAP SACC UK partners. The delivery of a cluster RCT, with process and economic evaluations, was feasible. The data indicate that NAP SACC UK has promise to increase PA levels of children aged 2–4 years during their time at nursery. Collectively, the findings from this study support the progression to a phase 3 trial of NAP SACC UK. A multicentre cluster randomised controlled trial to evaluate the effectiveness and cost-effectiveness of NAP SACC UK has been funded by NIHR (and will start in July 2019).