Author's response to reviews

Author's response to reviews Title: Effectiveness of a multi-strategy intervention in increasing the implementation of vegetable and fruit breaks by Australian primary schools: a non-randomised controlled trial. Authors: Nicole Nathan (Nicole.Nathan@hnehealth.nsw.gov.au) Luke Wolfenden (Luke.Wolfenden@hnehealth.nsw.gov.au) Andrew C Bell (Colin.Bell@newcastle.edu.au) Rebecca Wyse (Rebecca.Wyse@hnehealth.nsw.gov.au) Philip J Morgan (Philip.Morgan@newcastle.edu.au) Michelle Butler (Michelle.Butler@hnehealth.nsw.gov.au) Rachel Sutherland (Rachel.Sutherland@hnehealth.nsw.gov.au) Andrew J Milat (amila@doh.health.nsw.gov.au) Debra Hector (debra.hector@sydney.edu.au) John Wiggers (John.Wiggers@hnehealth.nsw.gov.au) Version: 2 Date: 6 July 2012 Author's response to reviews: see over

Hunter New England Population Health Direct Contact Details Phone: (02) 49246257 Fax: (02) 49246209 Email: Nicole.Nathan@hnehealth.nsw.gov.au 4 th July 2012 To the Editor RE: Effectiveness of a multi-strategy intervention in increasing the implementation of vegetable and fruit breaks by Australian primary schools: a non-randomised controlled trial. Thank you for the opportunity to respond to the reviewers comments and revise the manuscript. We have made a number of amendments based on the reviewers comments which we have described below: Reviewer: 1 Major Compulsory Revisions 1. Introduction- Please clarify that there are many types of FV programs and school fruit schemes, but that in this article the focus is on breaks during lessons and where FV is brought from home. We have included a paragraph (see below) in the introduction to acknowledge the different types of school fruit schemes and to define that our vegetable and fruit break is a time in class where children bring the food from home (lines 81-87 and lines 122-124). Consequently, a number of vegetable and fruit programs and schemes have been developed for implementation in schools. For example the United Kingdom [9], United States [21], Norway [22] and New Zealand [23] have school vegetable and fruit schemes which provide all or some students with a fully or partially subsidised piece of vegetable or fruit to consume each day at school. As an alternative to such distribution programs, Australian schools have been encouraged to implement a vegetable and fruit break program that provides a time in class for children to consume a piece of vegetable or fruit they have brought from home [24]. 2. Discussion- Please consider omitting the last part of the last sentence of the first paragraph in the discussion, because the study did not assess whether there was an increase in the children s intake. This sentence has been removed and the paragraph now re-written (see below) to reflect the focus of the study that is, being able to work with a large number of schools across a diverse geographic region (lines 311-313). These findings provide support for the use of a multi-strategic intervention approach in facilitating a large number of schools to implement vegetable and fruit programs across a large and diverse geographic region. Page 1

3. Discussion- I would like to read some reflections upon the following issues: What could the implementation of this intervention within broader programs ( Good for Kids, Good for life and Go for 2&5 ) have meant for the implementation intervention and the results? We have acknowledged in the discussion (see below) that as the intervention was part of a broader program this may have pre-disposed schools to adoption of obesity prevention programs. Thus, in the absence of such a program the potential effect size may have been smaller (lines 342-346). The intervention was undertaken as part of a larger multi-setting child obesity prevention program [32]. The existence of the broader program may have enhanced the salience of obesity prevention efforts in the intervention community and pre-disposed schools to the implementation of obesity prevention programs such as vegetable and fruit breaks. How could the data collection method (telephone interviews) and the broad answer categories have influenced the results? The manuscript has been amended (line 346-350.)(see below) to indicate the potential for over estimation of the intervention effect size due to social desirability response bias: Furthermore, the manuscript now suggests future studies should collect more objective measures of trial outcomes, or use survey modalities less subject to the influence of socially desirable responses such as web based surveys. Nonetheless, we believe the potential risk of telephone survey to the internal validity of this trial is minimal. In an unpublished study conducted by the authors, reporting of vegetable and fruit breaks by school Principals during a telephone survey was consistent with the reporting by preservice teachers at their school (Kappa = 1.0). The trial was also reliant on the self-reported existence of vegetable and fruit breaks collected as part of a telephone survey. The use of alternative survey modalities, such as the web may reduce the risk of social desirable responding [57], however, the use of direct observation would represent a more objective assessment of program adoption. How could delegation of reporting from the headmaster to someone else have influenced the results? Delegation of reporting of in this study was quite uncommon (88% were Principals or Acting Principals). The likelihood that systematic bias in the reporting of vegetable and fruit breaks between Principals and non-principals in this trial (even if it were to exist) influenced the trial findings is, therefore, remote. However, we have amended the discussion to acknowledge the report of Principals or their delegate may be biased and suggest future studies consider more objective measurement of vegetable and fruit break adoption (lines 346-350). The trial was also reliant on the self-reported existence of vegetable and fruit breaks collected as part of a telephone survey. The use of alternative survey modalities, such as the web may reduce the risk of social desirable responding [57], however, the use of direct observation would represent a more objective assessment of program adoption. Does having organized time for breaks automatically make the parents provide FV to bring and do pupils eat this FV? Page 2

We have amended the discussion to acknowledge, that whilst schools were required to identify strategies to ensure that all children participated, the existence of the program in the school does not necessarily result in 100% student participation (lines 354-359). Finally, the intervention encouraged schools to identify strategies to ensure that all children participated in the vegetable and fruit break. While data regarding student participation was not collected in the trial, anecdotally, Principal s frequently reported school wide participation by all students through school provision of vegetable or fruits to the small number of children unable to bring these foods from home, or by cutting up the vegetables and fruit provided by students and sharing it with all class members. 4. Table 4 The varying N s and high % are confusing. Is it possible to find a better way to present this so it is clearer what proportion of the schools took part in what? The table is challenging to present as the variables reported in it have been sourced from different data collection processes and hence the denominator changes between variables. To aid interpretation we have now included the denominator in each of the cells and kept the footnote with each to explain why the denominator changes. We believe this improves the clarity of information presented in the table. Minor Essential Revisions 5. Delivery of the multi-strategy intervention please explain what CATI (support calls) means. We have amended the wording in this sentence to make it clear that this is the 15-minute Computer-Assisted Telephone Interview (CATI) support call as explained in the intervention delivery table (lines 224-226). 6. Results/Table 3- consider including a subgroup analyses on who responded (headmaster vs others)? We are reluctant to include such analyses in the manuscript as the majority of participants were Principals or acting Principals (88%) and so cell sizes for a non Principal group are small and unable to detect meaningful between group differences. 7. Table 1 What was the content and application of the vegetable and fruit break policy? To support establish and sustain the program, schools were encouraged to implement a vegetable and fruit break policy which outlined how the program will be implemented, strategies the school will use to ensure all children participate and how the school will monitor the program. This detail has been included in Table 1 (see below). Individual school specific vegetable and fruit break policy- To support, establish and sustain the program schools developed a vegetable and fruit break policy outlining how the program will be implemented and monitored and strategies for ensuring no child misses out due to financial difficulties. Page 3

It is unclear whether the 1-day teacher relief funding was provided to the school champion for the 1-day workshop or this was something for additional staff. If it was the latter, was there a limitation of staff per school who could receive this? Table 1 has been amended (see below) to explain that this funding was for school champions to attend the workshop. 1-day (6 hour) healthy eating workshop for school champions held across the Hunter New England region. Fifteen workshops were held across the region to allow maximum access by schools. It is unclear whether the curriculum resources were developed for the implementation intervention or were the same as for the control schools. Please clarify here and in the manuscript. Table 1 has been amended to indicate that the curriculum material provided to intervention schools was the same as the materials control schools had access to. However, the other resources i.e. policy templates, information for parents, and newsletter articles were developed or modified specifically for intervention schools.. This detail has been included in Table 1 (see below). Resources were provided to schools: an easy-to-follow manual and a CD containing curriculum material (the same as that available for comparison schools), policy templates, information for parents, and newsletter articles (available for do wnload http://www.goodforkids.nsw.gov.au/parents). It is unclear how the telephone advice was tailored to the schools. Please elaborate here or in the manuscript. If during the scripted 15 minute CATI it was determined the school required more specific support the school was offered an additional support call from a trained health promotion support officer. The support officer then called the school contact to further discuss the barriers identified in the CATI and to offer advice or resources to meet the needs of that school. For example, during the 15 minute CATI some schools, located in low socioeconomic areas, identified that they needed advice as to how they can support students who can not afford to bring food from home. Knowing this information and the characteristics of the school, the support officer could then tailor the advice to the school. For example if the school was a Government school they let them know about grants that their school can apply for, how to do this and any support they may need to apply for the grant. This detail has been included in Table 1 (see below). If during the scripted 15 minute CATI it was determined the school required more specific support the school was offered additional support calls from a trained health promotion support officer. Schools received a maximum of two phone calls over a three-month period. The support officer called the school contact to further discuss the barriers identified in the CATI and to offer advice or resources to meet the needs of that school. In the last bullet point it is not clear who their schools refer to until the whole point is read. Please consider rewriting the bullet point. Page 4

This point has been re-written to place ownership at the front of the sentence (see below). Regional Directors and School Education Directors were provided two six-monthly reports, which described the proportions of their schools that participated in the workshops and were certified for vegetable and fruit breaks. Directors were asked to disseminate these results to Principals through existing communication channels such as Principal meetings and newsletters. 8. Table 2 and 3 The exact p-values are given in the column, but yet the * says alpha =0.05. It seems like the * could be removed. The asterisk was included to easily identify those results which were significant at the <0.05 level. Discretionary Revisions 9. Sample and recruitment (paragraph 2) the reference to figure 1 can be omitted here. 10. Figure 1 Consider including the response rate (%) from the result section in the figure, at least for those included in the analysis. 11. Table 3 Consider defining vegetable and fruit break in a footnote (i.e. scheduled class time to eat FV brought from home). Minor issues not for publication 12. Table 2 & 3 Consider moving references to footnotes. Given the analysis section references ARIA and SEIFA, we have removed these from the tables 13. Line 166 email should be e-mail 14. Line 198 and or should be and/or 15. Line 201 a period is missing at the end of the sentence Page 5

16. Line 324 Such is written twice, choose one of them. Reviewer: 2 sentence now reads that risk of such bias is likely to be limited 1. Overall, the manuscript addresses an important public health issue related low fruit and vegetables consumption among children. The research questions were clearly defined. The methods were appropriate and well described. Data were sound and presented appropriately. However, the authors should have a more thorough literature review regarding other randomized controlled trial on school-based fruit and vegetable promotion programs. Certainly, this is not the first school-based FV promotion controlled trial as the authors have claimed. Delgado-Noguera et al. Primary school interventions to promote fruit and vegetable consumption: a systematic review and meta-analysis, Prev Med. 2011. 53(1-2):3-9. Ramona Robinson-O'Brien et al. Impact of Garden-Based Youth Nutrition Intervention Programs: A Review. Journal of the American Dietetic Association. 2009, 109 (2),273 280 He, et al. Impact evaluation of the Northern Fruit and Vegetable Pilot Programme - a cluster-randomised controlled trial. Public Health Nutr. 2009 12(11):2199-208 We acknowledge that there is solid base of published randomised controlled trials on vegetable and fruit interventions in primary schools and consequently a rich history of systematic reviews [refs] published between 2000 and 2011. We thank the reviewer for the additional references of systematic reviews, which further supports the efficacy of school-based vegetable and fruit programs at increasing student consumption of vegetables and fruit. The primary outcome of interest, of the studies included within such systematic reviews however, is child vegetable and fruit intake, using such measures as 24hr dietary recall, plate waste or observation. Whilst our study involves a vegetable and fruit program, the focus of our research is adoption of such programs as opposed to their efficacy. As such, our outcome of interest is existence of the vegetable and fruit program within the school as opposed to whether the program increased student vegetable and fruit intake. To help clarify this difference we have clearly differentiated the efficacy of such programs, and the need for studies of strategies to aid their adoption in the background section (lines 90-101) Whilst there have been numerous calls for more applied research in schools and community settings, given the complexities of working in schools there is limited implementation research. The systematic review of implementation and dissemination interventions in community settings conducted by Rabin et al, which is referenced in the paper, failed to identify any implementation studies primarily focused on promoting child vegetable and fruit intake. As such, we believe that the limitations of implementation research noted within the Introduction section of the paper are valid. Nonetheless, we have included further detail into the introduction to qualify these statements. There have been calls for rigorous implementation research to identify the strategies that are effective in supporting population-wide implementation of such programs so that they benefit the health of the community [25]. The need for such research is illustrated by the findings of a recent update of an Agency for Health Care Policy and Research review published in 2010. The review of interventions in community based setting, including schools, primarily sought to evaluate the effectiveness of implementation or dissemination interventions on policies or programs promoting healthy eating, physical activity, sun protection, or prevent tobacco use. Implementation and dissemination studies were defined as those that evaluated the effectiveness of efforts that enable the widespread use of an evidence-based intervention by the target population and its successful integration Page 6

within a particular setting. The review identified just four school based interventions to improve the adoption of healthy eating policies or practices in schools and none targeted the adoption of vegetable and fruit initiatives specifically [26]. 2. In addition, the endpoint of this study was principals self-reported status of FV breaks. There were two potential limitations here: 1) principals' social desirability to report an ideal answer as the result of participating in the program; We have amended the discussion to acknowledge that the report of Principals or their delegate may be biased and suggest future studies consider more objective measurement of vegetable and fruit break adoption (lines 346-350). Delegation of reporting in this study was quite uncommon (88% were Principals or Acting Principals). The likelihood that systematic bias in the reporting of vegetable and fruit breaks between Principals and non-principals in this trial (even if it were to exist) influenced the trial findings is, therefore, remote. The trial was also reliant on the self-reported existence of vegetable and fruit breaks collected as part of a telephone survey. The use of alternative survey modalities, such as the web may reduce the risk of social desirable responding [57], however, the use of direct observation would represent a more objective assessment of program adoption. 2) Students may not bring and consume FV during the FV breaks. We have amended the discussion to acknowledge, that whilst schools were required to identify strategies to ensure that all children participate, the existence of the program in the school does not necessarily result in student participation (lines 354-359). Finally, the intervention encouraged schools to identify strategies to ensure that all children participated in the vegetable and fruit break. While data regarding student participation was not collected in the trial, anecdotally, Principal s frequently reported school wide participation by all students through school provision of vegetable or fruits to the small number of children unable to bring these foods from home, or by cutting up the vegetables and fruit provided by students and sharing it with all class members. We hope the revised manuscript meets your approval and look forward to receiving your response. Regards Ms Nicole Nathan Program Manager Hunter New England Population Health Locked Bag No. 10 Wallsend NSW 2287 AUSTRALIA Tel: +61 2 49246 257 Fax: +61 2 4924 6209 Email: Nicole.Nathan@hnehealth.nsw.gov.au Page 7