Research ArticleDigital Education to Limit Salt in the Home (DELISH)Program Improves Knowledge, Self-Efficacy, and BehaviorsAmong ChildrenCarley Ann Grimes, PhD; Alison Booth, PhD; Durreajam Khokhar, MS;Madeline West, BSc (Hons); Claire Margerison, PhD; Karen Jane Campbell, PhD;Caryl Anne Nowson, PhD

ABSTRACTObjective: To determine the efficacy of a Web-based salt reduction program on children’s salt-relatedknowledge, attitudes, and behaviors (KABs), self-efficacy, and intake of dietary salt.Design: Pretest and posttest. An online survey determined KABs and self-efficacy and a 24-hour urinecollection revealed salt intake.Setting: Victoria, Australia.Participants: Child–parent dyads (n = 102) recruited from 5 government schools.Intervention: A 5-week behavior-based education program delivered via weekly online interactive ed-ucation sessions.Main Outcome Measures: Change in KABs, self-efficacy, and daily salt intake.Analysis: Changes in outcomes were assessed using McNemar test, paired t test, and Cohen’s δ (CD).Results: A total of 83 children participated (mean age, 9.2 years [SD, 0.8 years]; 59% girls); 35% to 76%of children viewed weekly education session. Children with complete survey data (n = 75) had improvedscores for salt-related knowledge (+3.6 ± 0.4 points; P < .001; CD: 1.16), behaviors (+1.3 ± 0.1 points;P < .001; CD: 1.08), and self-efficacy (+0.9 ± 0.2 points; P < .001; CD: 0.64), but not attitude. Childrenwith valid urine collections (n = 51) showed no change in salt intake.Conclusions and Implications: Participation resulted in improvement of salt related knowledge, self-efficacy and behavior. Further research is required to confirm these results using a more robust study designwhich includes a control group. In addition, the long term impact on children’s salt intakes of comparableeducation programs needs to be assessed.Key Words: behavior, child, dietary sodium, education, Australia (J Nutr Educ Behav. 2018;50:547–554.)

Accepted April 3, 2018.

INTRODUCTION

Similar to those in the US,1 Austra-lian schoolchildren eat more dietarysalt (average intake of 6.1 g/d) than isrecommended for good health.2,3 Ahigher intake of salt during child-hood is associated with cardiovascularrisk factors.4,5 The presence of these risk

factors in early life is significantbecause both have been shown tofollow a tracking pattern over the lifespan6,7 as well as predict subclinical car-diovascular damage in adulthood.8,9

Additional concerns related to intakeof salty foods during early life in-cludes the development of tastepreferences that favor foods rich in salt,

which may lead to greater lifetimeintakes of salt.10

To protect cardiovascular health, in-terventions are required that seek toreduce salt in children’s diets. Al-though product reformulation oflower-sodium foods is considered anintegral component of population-wide salt reduction strategies,11 thesuccess of such initiatives is likely tobe maximized if there is also consum-er support for the availability of lower-salt foods.12 As such, education-basedprograms that raise awareness relatedto high salt intakes and food sourcesof salt are a fundamental compo-nent of population salt reductionprograms. Furthermore, evidenceamong adults indicates that the use ofbehavior-based strategies13 such as

Institute for Physical Activity and Nutrition, Deakin University, Geelong, AustraliaConflict of Interest Disclosure: The authors have not stated any conflicts of interest.Address for correspondence: Carley Ann Grimes, PhD, Institute for Physical Activity andNutrition, Deakin University, Locked Bag 20000, Waurn Ponds, Geelong, Victoria 3220,Australia; Phone: +61 3 9246223; Fax: + 61392446017; E-mail: carley.grimes@deakin.edu.au© 2018 Society for Nutrition Education and Behavior. Published by Elsevier, Inc. All rightsreserved.https://doi.org/10.1016/j.jneb.2018.04.002

Journal of Nutrition Education and Behavior • Volume 50, Number 6, 2018 547

Podcast available onlineat www.jneb.org

UNDER EMBARGO UNTIL JUNE 7, 2018, 12:01 AM ET

reading food labels to select lower-salt foods14 can reduce salt intake.Currently, less is known regarding theeffectiveness of education- andbehavior-based strategies to lower saltintakes among children. Accordingly,the authors developed a behavior-based salt reduction education program(Digital Education to Limit Salt in theHome [DELISH]) targeting elementaryschoolchildren in grades 2–4 (aged7–10 years) and their parents.15 Theprogram was Web-based because onlinetechnologies provide an important op-portunity to reach and engage withchildren16 and have been shown to bean effective medium to deliver inter-ventions seeking to alter children’sdietary behaviors.17 The aim of thisstudy was to determine the efficacy ofa Web-based salt reduction program onchildren’s salt-related knowledge, at-titudes, and behaviors (KABs), self-efficacy, and intake of dietary salt.

METHODS

Detailed information on the method-ology is reported in the studyprotocol.15 This was a single-arm studywith a pretest–posttest design in whichchildren completed surveys on salt-related KABs and self-efficacy, togetherwith 24-hour urine collections to assesschange in salt intake. The study waspowered (90% at α < .05) to detect a20% reduction in daily salt intake (ie,1.2 g/d), calculated using MLPowSimsoftware (version 1, University ofBristol, Bristol, UK, 2009). To accountfor a 20% dropout rate, the authorsaimed to recruit 122 children across6 schools.15 Children of varying so-cioeconomic levels were recruited fromprimary schools, which were locatedin the Greater Geelong region of Vic-toria, Australia. Ethics was approvedby the Deakin University Health andHuman Ethics Advisory Group (ProjectNo. HEAG-H 37/2016) and the De-partment of Education and Training,Victoria State Government (Project No.2015_002884). Written consent wasprovided by the school principal andthe primary caregiver; the child gavewritten assent.

Intervention Overview

Three behavioral messages weremodeled15 to meet the targeted 20%

reduction in children’s salt intake: (1)stop using the salt shaker, (2) switchto lower-salt foods by checking foodlabels (focus on bread, breakfast cereal,and cheese), and (3) swap processedsalty foods (eg, processed meats andtakeout and snack foods) withhealthier, lower-salt alternatives. Foodgroups that were previously identi-fied as important sources of salt in thediets of Australian children18 weretargeted.15 Full details of dietarymodeling performed to indicate thata 20% reduction in salt intake was fea-sible with the proposed interventionstrategies can be found elsewhere.15

The intervention content was embed-ded within the evidence-based healthyeating tool, the Australian Guide toHealthy Eating19; this served as a toolto communicate salt-specific mes-sages as well as place content withina wider healthy eating paradigm. In-tervention objectives, content, andstrategies were mapped to constructsof Social Cognitive Theory and be-havior change techniques.15 The5-week intervention included weeklyWeb-based interactive education ses-sions designed to take about 20minutes to complete. Sessions were de-veloped using the electronic learningsoftware Articulate Storyline 2 (Artic-ulate Global, Inc, New York, NY, 2015).Sessions were pilot-tested with 19 chil-dren aged 8–12 years and werereviewed for developmental appropri-ateness and comprehension by aprimary schoolteacher. A detailed over-view of information presented in theonline sessions is reported in the studyprotocol.15 In brief, a detective themenarrative that included comic strips atthe start of each session was used todeliver content (Figure 1). Childrenwere assigned a weekly detective casefile to solve that focused on the keyintervention behavioral messages. Ses-sions were designed to be fun, withinteractive games and activities, videocontent, and sound effects. Goal-setting activities related to the 3behavioral messages were included atthe end of sessions in weeks 2–4, towhich the children reported on theirprogress at the session the followingweek. A hard-copy detective logbookwas provided to track their goals andencourage discussion with theirparents about what they had learned.Children were rewarded with fun-

themed badges for solving case filesand meeting weekly goals. Finally, astudy website was accessible withaccess to intervention materials.Because parents are the gatekeepers offood for children of this age group, theintervention included parental re-sources (eg, online newsletters, shortmessage service texts).15 Parental out-comes are not the focus of this article.

Measures

Data collection was completed pre-post intervention. This includedcompletion of online surveys and an-thropometry and 24-hour urinecollections which were completed onsite at participating schools by trainedresearch staff.15 The child’s height andweight were measured using stan-dard protocols20 and a calibratedstadiometer (Model 220, Seca,Hamburg, Germany) and calibratedscales (UC-321 portable electronic scale,A&D Medical, San Jose, CA). Accord-ing to their measured height andweight, children were grouped intoweight categories using the Interna-tional Obesity Task Force body massindex reference cutoffs.21 Demograph-ics were reported at baseline by theprimary carer via an online question-naire; socioeconomic status was definedby parental educational attainment.

Urine collection. The 24-hour urine col-lection could be completed on eithera school or weekend day. When, pos-sible, children were encouraged tocomplete each measure on the sametype of day. The 24-hour urine collec-tion commenced with an emptybladder. Children were instructed tocollect all urine passed during 24 hoursinto a 2.5-L bottle.15 The collection wasfinished by providing a final void asclose as possible to the 24-hour finishtime. Children reported any missedcollections or spills. Children receivedverbal and pictorial written instruc-tions regarding how to complete the24-hour collection. Urinalysis wascompleted by an accredited commer-cial pathology laboratory (DorevitchPathology, Melbourne, Australia). Thetotal volume of the sample was re-corded and analyzed for sodiumconcentration using ion-selective elec-trodes and the Jaffe reaction forcreatinine concentration employing a

548 Grimes et al Journal of Nutrition Education and Behavior • Volume 50, Number 6, 2018

Siemens Advia 2400 analyzer (SiemensHealthcare Pty Ltd, Victoria, Austra-lia). The molecular weight of sodium(23 g/mol) and sodium chloride(58.5 g/mol) were used to convert lab-oratory values of millimoles to grams.Total volume, sodium, and creati-nine were standardized to a 24-hourperiod. Criteria were applied todetermine whether 24-hour urinesamples were valid.15

Online Salt Survey

Children completed a 29-item onlinesurvey related to dietary salt, whichcovered the constructs of knowledge(20 questions), attitudes (2 questions),behaviors (4 questions), and self-efficacy (3 questions). Questionnairedevelopment and scoring are reportedelsewhere.15 A copy of the question-naire is available upon request fromthe corresponding author. Test-retestreliability was assessed in a separatesample of 49 children (mean age, 9.0years; SD, 1.4 years; 61% male). Overall,the instrument displayed moderate togood test-retest reliability for con-

struct scores (ie, intraclass correlationcoefficients: knowledge, 0.66; atti-tude, 0.47; behavior, 0.70; and self-efficacy, 0.75)22 and poor to substantialagreement for individual survey items(ie, kappa coefficients for survey items,0.03–0.79)23 (Supplementary Data).Knowledge questions covered both de-clarative (eg, food sources of salt) andprocedural content (eg, reading foodlabels to pick foods with less salt). Re-sponses were in multiple choice formator true–false options. Correct answerswere scored 1 point with a maximumof 20 for knowledge construct. Atti-tude questions assessed the importanceof salt in food to make it tasty. The totalscore scale ranged from 0 to 6 points,with higher scores reflecting strongerbeliefs about the importance of salt fortaste. Two behavior questions relatedto discretionary salt use (ie, salt use atthe table and the availability of a saltshaker on the table during mealtimes).Specifically, children were asked, Do youadd salt to your meal at the table? andIs a salt shaker placed on your table athome during mealtimes? Both ques-tions were previously used with

children as a measure of discretion-ary salt use.24 Two more questionsrelated to other targeted behaviors inthe intervention (ie, In the last month,did you speak to your parents about buyingfoods which have less salt? and In the lastmonth, did you speak to your parents aboutthe use of the salt shaker in your home?).Potential behavior scores ranged from0 to 8 points, with higher scores in-dicating better adherence to targetedsalt-related behaviors. Finally, self-efficacy questions related to children’sbelief that they could perform specif-ic behaviors targeted within theintervention, such as their perceivedability to (1) swap lunchtime foods tothose with less salt, (2) read food labelsto pick foods with less salt, and (3)reduce table salt use. Potential self-efficacy scores ranged from 0 to 6points, with higher scores indicatinga greater degree of self-efficacy to lowerthe amount of salt in the diet.

Process Evaluation

Intervention delivery was assessedusing metric data15 collected within the

Figure 1. Examples of content from the Web-based interactive education sessions completed by children.

Journal of Nutrition Education and Behavior • Volume 50, Number 6, 2018 Grimes et al 549

electronic learning software (Articu-late Online) used to deliver Websessions. After the intervention, the re-searchers assessed the acceptability ofthe program via an anonymous child-reported online evaluation survey.

Data Analysis

Analysis was completed using StataSE (version 14, StataCorp LP, CollegeStation, TX, 2015). The researchersused descriptive statistics to describecategorical variables (n, %) and con-tinuous variables (mean and SD or SE).Normality of variables was determinedvia histogram distributions. To assessthe change in the proportion of chil-dren who reported correct or incorrectresponses on the salt survey before andafter the intervention, the research-ers used the exact McNemar test.Owing to multiple testing for knowl-edge items (20 questions), Bonferronicorrection was used to adjust P to a sig-nificance level of .05 / 20 = .003. Forall other analyses, 2-tailed P < .05 wasconsidered significant. To assess changein survey construct scores, a paired ttest was used and Cohen’s δ was cal-culated to determine the effect size.

The proportion of children with saltintake exceeding the age-specific upperlevel (UL) for sodium intake was cal-culated: ie, 1,400 mg/d (salt equivalent,3.5 g/d) for 4- to 8-year-olds and2,000 mg/d (salt equivalent, 5.0 g/d)for 9- to 12-year-olds.3 Change in saltintake (grams per day) was assessedusing a paired t test and change in pro-portion exceeding the UL was assessedusing the exact McNemar test.

RESULTS

Figure 2 shows participant recruit-ment and retention. A total of 83children completed the intervention(average age, 9.2 years; SD, 0.8 years;n = 49 [59%] female). Most (n = 69[83%]) were of a healthy weight andthere was a reasonable spread of so-cioeconomic backgrounds (ie, n = 21[27%], 32 [32%], and 35 [43%], low,middle, and high socioeconomicstatus, respectively).

Knowledge, attitudes, behaviors andself-efficacy related to salt intake. Of the83 children who completed the inter-vention, 75 also completed a pre-post salt survey (Figure 2). Among

these children, after participation therewas significant improvement in chil-dren’s overall knowledge, behavior,and self-efficacy scores (all P < .001)and no change in attitude score (P =.075) (Table 1). Of the 20 individualknowledge items assessed, improve-ments were found in 8 items(Supplementary Data), which indi-cated improvements in declarative andprocedural knowledge. For example,after the intervention, more childrenwere aware of the recommended dailysalt intake (n = 33 [44%] pre vs n = 54[72%] post; P = .001) and could cor-rectly identify checking the sodiumcontent on the food label as a strate-gy to cut down on salt (n = 39 [52%]pre vs n = 68 [91%] post; P < .001).Overall, children’s knowledge relatedto food sources of salt improved.

There was no shift in the propor-tion of children who reported that saltmakes food tasty (agree a lot/agree wasn = 45 [60%] pre vs n = 44 [59%] post;P = .999) or that low salt food tastesbad (agree a lot/agree was n = 10 [13%]pre vs n = 9 [12%] post; P = .999).Overall, children reported positivechanges in salt-related behaviors(Figure 3). There was a significant 19%

Figure 2. Flowchart of participant recruitment and retention within the Digital Education to Limit Salt in the Home program.Data do not equal the number of incomplete collections because a urine sample can meet more than 1 criterion. RR indi-cates response rate; SES, socioeconomic status.

550 Grimes et al Journal of Nutrition Education and Behavior • Volume 50, Number 6, 2018

reduction in the proportion of chil-dren who reported that a salt shakerwas placed on the table during meal-times (P = .002). The 8% reduction inreported use of table salt did not reachstatistical significance (P = .237). Whenthe salt shaker was not placed on thetable, fewer children (26%) reportedthat they added salt (Pearson χ2 = 17.7;P < .001), compared with 70% who re-ported adding salt when a salt shakerwas on the table (the direction of as-sociation was the same before and afterthe intervention). Postintervention,significantly more children reportedthat they had spoken to their parentsabout using the salt shaker in thehome and buying salt-reduced foods(Figure 3). There was also improve-ment in the children’s own reportedbelief that they could change their be-haviors to eat less salt (SupplementaryData).

Change in salt intake. Of the 83 chil-dren who completed the program, 51provided a valid pre-post urine sample(Figure 2). Among these children, therewas no change in salt intake frombefore (5.4 ± 0.4 g/d) to after the in-tervention (5.3 ± 0.3 g/d) (P = .751).There was no change in the propor-tion of children who exceeded the ULfor salt intake before (n = 31; 61%) andafter the intervention (n = 34; 67%)(exact McNemar test, P = .607).

Process evaluation. Most childrenviewed the majority of weekly Web ses-sions; three quarters viewing the week1 session and about 60% viewed ses-sions at weeks 2–4, except for the finalweek 5 recap session, which was viewedby only a third of children. Among chil-dren who viewed sessions, completionrates were relatively high. The averageduration to complete sessions was

23 minutes (Supplementary Data).Findings from the online evaluationsurvey, which was completed by 59 ofthe 83 children who completed theprogram (ie, a 71% response rate), in-dicated that overall, the program waswell received, most children reportedthat they liked completing the program(n = 53; 90%), and that it was inter-esting (n = 48; 81%) (SupplementaryData). Some individual comments in-cluded “I loved to do it [the program]after school,” and “It was very inter-esting and fun. I will tell my friendsall about this.” Aspects of the Web ses-sions that children reported liking mostincluded games and activities, the dogprotagonist, and comics. A minorityof children (n = 6; 10%) reported somefactors that they did not like (eg, thelength of computer sessions, no textvoice-over available). About three quar-ters of children reported that theythought the program helped them toeat less salt (Supplementary Data). Most(n = 41; 70%) could recall somethingthey had learned from the program.The most common response related toreading food labels to find foods withless salt (n = 11; 19%). Some (n = 14;24%) reported experiencing a com-puter problem while watching sessions(eg, frozen session, improper buttonfunctionality).

In weeks 2–4, children who reachedat least 90% completion of the Web-based session were prompted to set agoal related to the coming week’s

Table 1. Change in average salt-related knowledge, attitudes, behavior andself-efficacy scores pre- and post-intervention (n = 75)

ConstructScore Pre Post Changea P-value

Effect size(Cohen’s δ)

Knowledge 13.5 ± 0.3 17.1 ± 0.3 3.6 ± 0.4 <0.001 1.16

Attitude 2.2 ± 0.1 2.0 ± 0.1 –0.2 ± 0.5 0.075 –0.15

Behavior 5.2 ± 0.1 6.5 ± 0.1 1.3 ± 0.1 <0.001 1.08

Self-efficacy 3.9 ± 0.2 4.7 ± 0.1 0.9 ± 0.2 <0.001 0.64

aChange assessed via paired t-test.

Figure 3. Proportion (%) of children who reported engaging in salt-related behaviors pre- and postintervention (n = 75). Forthe salt shaker question, n = 74 because 1 I’m not sure response was excluded from the analysis. Change in proportion pre-and postintervention was determined by exact McNemar test. *P = .002; **P < .001.

Journal of Nutrition Education and Behavior • Volume 50, Number 6, 2018 Grimes et al 551

behavioral message. About half of the83 children in the intervention setweekly goals (eg, n = 45 [54%], n = 42[51%], and n = 42) [51%] at weeks 2–4,respectively). At the start of the fol-lowing week’s Web session (ie, weeks3–5), the children were asked to reportwhether they had met their goal. Intotal, across each week, 52, 48, and 29children, respectively, reported on theirgoal outcome. Among these children,50 (96%) reported meeting the week2 goal related to discretionary salt use,36 (75%) reported meeting the week3 goal related to switching to lower-salt bread, cheese, and/or breakfastcereal, and 13 (45%) reported meetingthe week 4 goal related to swappingsalty processed foods with healthieralternatives.

DISCUSSION

The findings from this study demon-strate that participation in a Web-based salt reduction educationprogram (ie, DELISH) led to improve-ments in children’s salt-relatedknowledge, behaviors, and self-efficacy.However, there was no change in at-titudes related to the importance of saltto make food tasty, nor was there achange in daily salt intake. It appearsthat the content, which was linked tobehavioral theory and the medium ofdelivery (eg, online interactive ses-sions) was engaging for the targetedage group of children in school grades2–4.

Children’s improvement in salt-related knowledge was substantial, asindicated by the 1.16 SD distributionshift. Although specific to salt, theseresults are comparable to reportedimprovements in elementary school-children’s nutrition knowledge afterparticipation in school-based nutri-tion education interventions inthe US25 and UK.26 In the currentstudy, particular areas in which the re-searchers observed improvement inknowledge related to increased aware-ness of high salt intakes amongchildren, dietary recommendations forsalt, and food sources of salt. For mostof the knowledge items that showedno improvement, baseline levels ofknowledge were already high (>80%).Future studies using this question-naire should consider the itemdifficulty index of these questions and

modify the questions to ensure ade-quate difficulty. With regards to foodlabel related knowledge, some posi-tive changes were observed. Forexample, significantly more childrenknew to utilize the sodium content per100-g column for comparing saltcontent of food products (P < .001) inpost-intervention. There was some in-dication of an increase in theproportion of children (ie, pre: n = 37[49%] to post: n = 59 [79%]; P = .01)who could use the nutrition informa-tion to correctly identify a bread withless salt. However, it should be notedthat this was not considered signifi-cant using the more conservativeBonferroni correction for multiplecomparisons P value of <.003. Theformat in which intervention contentwas presented to children was de-signed to be fun and interactive,15 andit is likely that this contributed to therelatively high log-in rates for weeklyonline sessions, which in turn allowedfor favorable outcomes related to im-proved awareness of intake and sourcesof salt, and reading food labels.

After the intervention, children dis-played a greater degree of self-efficacyto perform behaviors to lower theamount of salt in their diet. In par-ticular, improvements were noted forthe ability to choose lower-salt foodsat lunchtime and read a food label tochoose foods with less salt, althoughthis was a self-report measure and foodavailability was not measured. It is pos-sible that the children’s increasedperceived ability to perform these tasksresulted from the inclusion of relatedpractice activities within Web-basedsessions.15 Conversely, there was noimprovement in children’s belief thatthey could stop adding salt at thetable. This was perhaps the simplestbehavior listed, with no specializedskill development required; it may alsoexplain the already high baseline valueof 97%.

Overall, there was a large improve-ment in children’s reported salt-relatedbehaviors. After the intervention, fewerchildren reported placing a salt shakeron the table; children who reportedthis behavior were also less likely toadd salt at the table. Hence, target-ing the availability of discretionary saltin the home and norms related tomeal practices may be important inshaping children’s salt use habits.

Furthermore these findings suggestthat simple education messages andbehavior-based strategies appear to beable to influence children’s discretion-ary salt use behaviors. Althoughdiscretionary salt contributes far lesssalt to the diet compared with pro-cessed foods,27 education regardingdiscretionary salt use practices canserve as a useful tool to raise aware-ness more broadly about salt in foodand put the topic of high salt intakeson parents’ radars. The interventionwas effective in encouraging chil-dren to talk to their parents aboutkey salt-related behaviors, such asbuying salt-reduced foods and remov-ing the salt shaker from the table.Because parents are the gatekeepers offood within the home, this a posi-tive outcome and speaks to thepotential for children to act as agentsof change in the home.28

Despite the improvements in salt-related behaviors, the researchers sawno change in salt intake as measuredby 24-hour urine collection, but thiswas expected because of the smallsample size: only 53 children provid-ed valid 24-hour urine collectionsbefore and after the intervention (wellbelow the target sample size of 102children). Although some other saltreduction programs achieved a signif-icant reduction in salt intake inprimary schoolchildren (−1.9 g/d inChina29 and −1.1 g/d in Portugal30),those studies provided salt substi-tutes and took place over a muchlonger period (about 3 months). Incontrast, the current intervention tookplace for 5 weeks and may have beentoo short to detect a move toward pur-chasing lower-salt manufacturedproducts.

There were a number of limita-tions within this study. First, theomission of a control group meansthat no causal inference can be made.The improved outcomes reported forknowledge, behaviors, and self-efficacymay have resulted from factors outsidethe intervention. Moderate reliabili-ty for the construct scores of KABsuggests imprecision in the survey in-strument and decreases the ability tomatch changes to the intervention.31

Findings from the 3 survey items thatshowed poor or slight agreement inreliability testing should also be in-terpreted with caution. The low kappa

552 Grimes et al Journal of Nutrition Education and Behavior • Volume 50, Number 6, 2018

values suggest that questions may havebeen unclear or too difficult for chil-dren of this age, resulting in guessing;future work should consider revisingthese items. The small sample size,combined with recruitment restrictedto 1 region of Victoria, limits thegeneralizability of the findings. Fur-thermore, results were limited tochildren with pretest and posttest mea-sures for the salt survey and childrenwho returned a complete 24-hoururine collection. Self-reported ques-tionnaire data for behaviors are subjectto potential social desirability bias. Astrength of the study was the use ofa questionnaire developed for the in-tervention, which showed overallacceptable test-retest reliability and useof 24-hour urine samples to measuresalt intake. Furthermore, the interven-tion content and strategies wererigorously developed with reference todietary modeling and behavioralchange theory.15

IMPLICATIONS FORRESEARCH ANDPRACTICE

Similar to previous studies conductedin Australia2 and the US,1 findings fromthis study indicate that salt intakeamong primary schoolchildren aged7–10 years is too high. This has adverseimplications for future cardiovascu-lar health outcomes.4 Although notspecific to salt, evidence suggests thatthe acquisition of nutrition knowl-edge more broadly is an importantprecursor for positive food-related be-havior change,32 and higher levels areassociated with improved dietaryintakes among children.33 Self-efficacyalso appears to be particularly impor-tant in shaping children’s nutrition-related behaviors,34 and it isencouraging that this construct wasable to be modified through interac-tive Web-based activities. Despiteimprovements in salt-related knowl-edge and self-efficacy, it may still bedifficult for families to lower salt intakebecause the addition of salt is so wide-spread across the food supply.35 Giventhe favorable outcomes for impor-tant mediators of behavior change(ie, knowledge and self-efficacy) re-ported in this study, combined withoverall positive process data and user

likeability, future research shouldexplore alternative means of dissemi-nating the program. In this pilot study,children were recruited via schools andthe program was delivered within thehome. Future work should explorewhether the program could be trans-lated into the school environment toenable wider reach. This would includescoping of current healthy lifestyle-related programs delivered in schoolsand understanding enablers and bar-riers to implementation. With thisknowledge, the DELISH program couldbe tailored to suit the school environ-ment. Future studies with follow-upassessment are also required to deter-mine whether these positive changesin salt-related knowledge, self-efficacy,and behaviors are sustained.

ACKNOWLEDGMENTS

This study was funded by theHeart Foundation (Vanguard GrantApplication 100574). During the de-velopment of the intervention, CAGwas supported by a Heart Founda-tion Postdoctoral Fellowship (Award100155). CAG is supported by anAlfred Deakin Postdoctoral Fellow-ship. The authors acknowledge theAustralian Department of Educationand Training for their support in al-lowing recruitment and data collectionto be conducted within the govern-ment school sector. They thank allparticipating schools and families.They wish to acknowledge BenPearmain Illustration for artwork usedin the intervention materials and con-tributing to the development of thenarrative. They also thank research as-sistants Carrie Service and MareeBouterakos for their contribution tothe development of intervention re-sources, and Anne Griffiths in her roleas project coordinator.

SUPPLEMENTARY DATA

Supplementary data related to thisarticle can be found at https://doi.org/10.1016/j.jneb.2018.04.002.

REFERENCES

1. Cogswell ME, Yuan K, Gunn JP, et al.Vital signs: sodium intake among U.S.school-aged children—2009–2010.

MMWR Morb Mortal Wkly Rep. 2014;63:789-797.

2. Grimes CA, Riddell LJ, Campbell KJ,et al. Dietary intake and sources ofsodium and potassium among Austra-lian schoolchildren: Salt and OtherNutrients in Children (SONIC) study. BMJOpen. 2017;7:e016639.

3. National Health and Medical ResearchCouncil. Nutrient reference values forAustralia and New Zealand. Update 1.1revision of sodium. 2017. https://www.nrv.gov.au/nutrients/sodium. AccessedNovember 20, 2017.

4. Appel LJ, Lichtenstein AH, Callahan EA,Sinaiko A, Van Horn L, Whitsel L. Re-ducing sodium intake in children: apublic health investment. J Clin Hypertens.2015;17:657-662.

5. Grimes CA, Riddell LJ, Campbell KJ,He FJ, Nowson CA. 24-h urinarysodium excretion is associated withobesity in a cross-sectional sample of Aus-tralian schoolchildren. Br J Nutr. 2016;115:1071-1079.

6. Chen X, Wang Y. Tracking of bloodpressure from childhood to adulthood:a systematic review and meta-regressionanalysis. Circulation. 2008;117:3171-3180.

7. Singh AS, Mulder C, Twisk JWR, vanMechelen W, Chinapaw MJM. Track-ing of childhood overweight intoadulthood: a systematic review of the lit-erature. Obes Rev. 2008;9:474-488.

8. Lai CC, Sun D, Cen R, et al. Impactof long-term burden of excessive adi-posity and elevated blood pressure fromchildhood on adulthood left ventricu-lar remodeling patterns: the BogalusaHeart Study. J Am Coll Cardiol. 2014;64:1580-1587.

9. Aatola H, Hutri-Kahonen N, Juonala M,et al. Lifetime risk factors and arterialpulse wave velocity in adulthood: thecardiovascular risk in young Finns study.Hypertension. 2010;55:806-811.

10. Stein L, Cowart B, Beauchamp G. Thedevelopment of salty taste acceptance isrelated to dietary experience in humaninfants: A prospective study. Am J ClinNutr. 2012;95:123-129.

11. Institute of Medicine. Strategies to ReduceSodium Intake in the United States. Wash-ington, DC: National Academy Press;2010.

12. Regan A, Kent MP, Raats MM,McConnon A, Wall P, Dubois L. Ap-plying a consumer behavior lens to saltreduction initiatives. Nutrients. 2017;9:901.

Journal of Nutrition Education and Behavior • Volume 50, Number 6, 2018 Grimes et al 553

13. Trieu K, McMahon E, Santos JA, et al.Review of behaviour change interven-tions to reduce population salt intake. IntJ Behav Nutr Phys Act. 2017;14:17.

14. Ireland DM, Clifton PM, Keogh JB.Achieving the salt intake target of 6 g/day in the current food supply in free-living adults using two dietary educationstrategies. J Am Diet Assoc. 2010;110:763-767.

15. Grimes CA, Booth A, Khokhar D, et al.The development of a Web-basedprogram to reduce dietary salt intake inschoolchildren: study protocol. JMIR ResProtoc. 2017;6:e103.

16. Li JS, Barnett TA, Goodman E, et al.Approaches to the prevention and man-agement of childhood obesity: the roleof social networks and the use of socialmedia and related electronic technologies:a scientific statement from the Ameri-can Heart Association. Circulation. 2013;127:260-267.

17. Cullen KW, Thompson D, Boushey C,Konzelmann K, Chen TA. Evaluationof a web-based program promotinghealthy eating and physical activity foradolescents: teen choice: food andfitness. Health Educ Res. 2013;28:704-714.

18. Grimes CA, Campbell KJ, Riddell LJ,Nowson CA. Sources of sodium in Aus-tralian children’s diets and the effect ofthe application of sodium targets to foodproducts to reduce sodium intake. Br JNutr. 2011;105:468-477.

19. National Health and Medical ResearchCouncil. 2013. Australian guide tohealthy eating. https://www.eatforhealth.gov.au/guidelines/australian-guide-healthy-eating. Accessed July 15,2016.

20. Marfell-Jones M, Olds T, Stewart A,Carter L. International Standards for An-thropometric Assessment. Potchefstroom,South Africa: International Society forthe Advancement of Kinanthropometry;2006.

21. Cole TJ. Establishing a standard defini-tion for child overweight and obesityworldwide: international survey. BMJ.2000;320:1-6.

22. Koo TK, Li MY. A guideline of select-ing and reporting intraclass correlationcoefficients for reliability research. JChiropr Med. 2016;15:155-163.

23. Landis JR, Koch GG. The measurementof observer agreement for categoricaldata. Biometrics. 1977;33:159-174.

24. Grimes CA, Baxter JR, Campbell KJ,et al. Cross-sectional study of 24-hoururinary electrolyte excretion and asso-ciated health outcomes in a conveniencesample of Australian primaryschoolchildren: the Salt and Other Nu-trients in Children (SONIC) studyprotocol. JMIR Res Protoc. 2015;4:e7.

25. Gower JR, Moyer-Mileur LJ, Wilkin-son RD, Slater H, Jordan KC. Validityand reliability of a nutrition knowl-edge survey for assessment in elementaryschool children. J Am Diet Assoc. 2010;110:452-456.

26. Lakshman RR, Sharp SJ, Ong KK,Forouhi NG. A novel school-basedintervention to improve nutrition knowl-edge in children: cluster randomisedcontrolled trial. BMC Public Health.2010;10:123.

27. James PT, Ralph A, Sanchez-Castillo CP.The dominance of salt in manufac-tured food in the sodium intake ofaffluent societies. Lancet. 1987;329:426-429.

28. Gunawardena N, Kurotani K,Indrawansa S, Nonaka D, Mizoue T,Samarasinghe D. School-based inter-vention to enable school children to actas change agents on weight, physical ac-tivity and diet of their mothers: a clusterrandomized controlled trial. Int J BehavNutr Phys Act. 2016;6:45.

29. He FJ, Wu YF, Feng XX, et al. Schoolbased education programme to reducesalt intake in children and their fami-lies (School-EduSalt): cluster randomisedcontrolled trial. BMJ. 2015;350:h770.

30. Cotter J, Cotter MJ, Oliveira P, CunhaP, Polónia J. Salt intake in children 10–12 years old and its modification by activeworking practices in a school garden. JHypertens. 2013;31:1966-1971.

31. Pallant J. SPSS Survival Manual: A Stepby Step Guide to Data Analysis Using IBMSPSS. Sydney, Australia: Allen & Unwin;2016.

32. Worsley A. Nutrition knowledge andfood consumption: can nutrition knowl-edge change food behaviour? Asia PacJ Clin Nutr. 2002;11(suppl 3):S579-S585.

33. Vaitkeviciute R, Ball LE, Harris N. Therelationship between food literacy anddietary intake in adolescents: a system-atic review. Public Health Nutr. 2015;18:649-658.

34. Hall E, Chai W, Albrecht JA. Relation-ships between nutrition-relatedknowledge, self-efficacy, and behaviorfor fifth grade students attending TitleI and non-Title I schools. Appetite. 2016;1:245-253.

35. Webster JL, Dunford EK, Neal BC. Asystematic survey of the sodium con-tents of processed foods. Am J Clin Nutr.2010;91:413-420.

554 Grimes et al Journal of Nutrition Education and Behavior • Volume 50, Number 6, 2018