Development and piloting of a primary school-based salt reduction programme: formative work and a process evaluation in rural and urban Malawi

Qualitative research, trial development, process evaluation , focus groups, salt 23 reduction, schools, Normalisation Process Theory 24 Word count = 5930, excluding title page, abstract, references, figures and tables. 25

300 words 26

27

Excess salt intake is a major modifiable risk factor for cardiovascular disease . 28 Promoting salt reduction as part of routine school -health programming may be a pragmatic 29 way to address this risk factor early in the life course but has not been tested in sub-Saharan 30 Africa (SSA). Here we describe the formative work with stakeholders and process evaluation 31 of pilot work to develop a school-based salt reduction programme for children aged 11-14 years, 32 in preparation for a cluster-randomised trial in rural/urban Malawi. 33

Collection of observational data and documentary evidence (meeting 34 minutes/field notes) from the earliest key stakeholder engagement with Malawi Ministries of 35 Health, Education, Local Government and Rural Development and Malawi Institute of 36 Education, and non-governmental stakeholders; and a series of semi-structured interviews and 37 focus groups (with head teachers (n=2); teachers (n=4); parents (n=30); and learners (n=40)) . 38 Data analysed thematically and conceptualised through a Normalization Process Theory lens. 39

Formative work illustrated a range of administrative, technical, and practical issues 40 faced during development of the programme ; including allocation of stakeholder roles and 41 responsibilities, harmonisation with pre-existing strategies and competing priorities, resources 42 required for programme development, and design of effective teaching materials. While 43 participants were positive about the programme, the process evaluation identified features to 44 be refined including perceived challenges to participation, recommended adaptations to the 45 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 3 content and delivery of lessons, and concerns related to quantity/quality of learning resources 46 provided. 47

This study demonstrates the importance of comprehensive, sustained , and 48 participatory stakeholder engagement in the development of a novel school health programme 49 in SSA; and highlights the factors that were critical to successfully achieving this. We also 50 demonstrate the value of detailed process evaluation in informing development of the 51 programme to ensure that it was feasible and relevant to the context prior to evaluation through 52 a cluster-randomised trial. 53

54 Non-communicable diseases (NCDs) are a growing public health concern in Low and Middle 55 Income Countries (LMICs) especially in sub-Saharan Africa (SSA).[1, 2] Hypertension is the 56 leading and most important risk factor for cardiovascular diseases (CVDs), which are a major 57 cause of death.[3] There is therefore a growing need for population-based interventions. 58 An important modifiable risk factor for hypertension is sodium intake,[4] which is 59 predominantly related to the level of salt in diets. [5] The World Health Organization (WHO) 60 has set recommended maximum daily intakes ; more than 2g per day of sodium intake 61 (equivalent to 5g of salt) risks development of hypertension.[6-8] Globally, estimated mean 62 sodium consumption is 3.95g, equivalent to 9.88g of salt per day.[9] Although previously the 63 highest sodium intake was seen in high income countries,[9] similar trends are now evident in 64 SSA. While specific statistics about salt intake in children in SSA are largely unavailable,[10] 65 higher sodium intake in urban rather than rural areas of SSA has been reported [11] among 66 both adults and children.[7, 12, 13] There is emerging evidence that dietary salt reduction can 67 reduce blood pressure.[5, 7] A meta-analysis of 10 trials in participants aged 8-16 years showed 68 that a 4-week modest salt intake reduced mean blood pressure by 1.2/1.3mmHg.[6, 14, 15] 69 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 4 The WHO’s global action plan for the prevention and control of NCDs 2013-2020 stipulates a 70 relative 30% reduction in salt intake as one of the ‘best buys’ of the nine global NCD targets.[16] 71 Malawi has a high discretionary salt intake and therefore individual or household salt reduction 72 has the potential to b e a highly cost -effective strategy for reducing hypertension and 73 cardiovascular diseases.[2] This evidence provides the rationale for the proposed trial of a salt 74 reduction intervention in schools in Malawi. 75 The ‘No to Sodium’ (No to Na ) trial aims to evaluate a 12-week salt reduction programme 76 embedded in the science and technology curriculum of primary school children (aged 11 -14) 77 through a cluster-randomized trial in rural (Karonga district) and urban (Lilongwe district) 78 Malawi. This trial builds on a previous trial evaluating the effectiveness of schools in reducing 79 the salt intake of children and their parents .[6, 15] In this article we describe formative work 80 done through a process of stakeholder engagement , lessons learnt from a pilot of the 81 programme and how these influenced the design of the final programme to be evaluated in the 82 main trial . We used Normalization Process Theory (NPT) as our underpinning theoretical 83 framework. NPT has been used extensively for designing and evaluating complex health 84 interventions[17] and has been shown to help trialists ‘describe, assess and enhance 85 implementation potential’.[17, 18] Our aims were: 86 1. To describe the formative work of stakeholder engagement undertaken to develop and 87 refine a salt reduction programme in Malawian primary schools. 88 2. To describe qualitative findings relating to perceptions of the programme, including 89 identified challenges and facilitators likely to influence success of implementation of 90 the programme during the trial. 91 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 5

92 Design 93 We describe the qualitative data collectio n methods employed to address our research aims 94 which involved formative work (Phase 1); and a process evaluation of the programme pilot 95 (Phase 2). 96 Planned trial 97 This study was conducted in preparation for evaluation of the programme through a cluster 98 randomized intervention trial in 11–14-year-old school attenders and their families in rural and 99 urban Malawi. The trial involves 26 schools, with thirteen delivering the 12-week programme 100 (consisting of lessons and competitions for children and other activities with parents) on salt 101 reduction to children and their family members as part of the routine curriculum , and thirteen 102 control schools follow ing the routine curriculum only. Full details of the trial are described 103 elsewhere. (https://doi.org/10.1186/ISRCTN13909759) 104 Ethics statement 105 Ethical approval was received from the National Committee on Research in the Social Sciences 106 and Humanities in Malawi (P.11/18/333) and the College of Medical , Veterinary and Life 107 Sciences Ethics Committee at the University of Glasgow. Verbal and written information about 108 the study were provided, in participants’ mother tongue prior to obtaining written informed 109 consent. All Focus Group Discussions (FGDs) with adolescents only took place after obtaining 110 written signed consent from parents and written signed assent from adolescents. This study is 111 registered at ISRCTN (#13909759). 112 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 6 Data collection 113 Phase 1 Formative work (preliminary engagement activities undertaken to 114 develop and refine a school-based salt reduction programme) 115 The formative work included engaging with local stakeholders in key ministries: Ministries of 116 Health (MoH), Education (MoE), Local Government and Rural Development (MLGRD) and 117 the Malawi Institute of Education (MIE), and non-governmental stakeholders (from public and 118 private sector including the business community) and adolescents and their guardians, to assess 119 the feasibility and acceptability of the programme before implementation of the trial (Figure 120 1). 121 122 The School Health and Nutrition (SHN) Programme, a joint programme between MoH and 123 MoE, had been engaged at grant proposal stage (through individual meetings and e-mails) and 124 had negotiated support letters for the application from MoH and MoE. On award of the funding, 125 further meetings took place with the ministries to generate an action pla n for stakeholder 126 engagement. The process of stakeholder engagement continued with a series of small meetings 127 between the study leads and key people in the MoH and MoE (Fig 1). These meetings involved 128 small groups (less than 5) and were held in their offic es. The study team helped stakeholders 129 understand the proposed trial by making comprehensive power point presentations at the 130 beginning of each meeting. The meetings explored views on the planned trial and provided the 131 opportunity to ask questions, enabling the stakeholders to get a clear understanding of the trial 132 and its objectives. 133 134 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 7 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 Fig 1. A pictorial chart of key stakeholders 153 Ministry of Health (MoH) Ministry of Education (MoE) Department of Clinical Services DNHA Secretary for Education (SE) SHN Director of Education, Youth and Sports BASIC DQAS DTED MIE Government SHN Coordinator Head Teachers Teachers Secretary for Health (SH) Parents Parent Teacher Association (PTA) Learners Ministry of Local Government and Rural Development (MLGRD) District Council Office (DC) District Executive Committee (DEC) NCD- MH & I HEU School Management Committee Mother group District Education Office (PEAs) Community Nursing Secretary for MLGRD . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 8 Data was collected through minutes of meetings and field notes, from the very earliest stages 154 of engagement which documented the extensive preliminary work undertaken (Table 1) from 155 when the outline proposal was developed through to when funding was received, and materials 156 were developed. This initial evaluation work explored how key stakeholders viewed the 157 proposed programme, and factors influencing these perspectives and responses. 158 159 160 161 162 163 164 165 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 9 Table 1. Showing stakeholder engagement meetings 166 TIMELINE ACTIVITY FIRST APPLICATION 8th December 2015 PI meeting with MoH Non- Communicable Diseases, mental health, and Injury (NCD- MH & I) department lead regarding application 12th December 2015 PI introduced proposed study to SHN leads through NCD-MH & I department lead at MoH (co-applicant) December 2015- January 2016 Calls and informal meetings with MoH and SNH leads on study 20th January 2016 Letter of Support MoH, Letter of Support MoE through SHN 21st January 2016 Outline proposal submission Outline proposal rejected SECOND APPLICATION May-June 2017 Calls and informal meetings MoH and SHN leads 15th June 2017 New outline proposal submitted with original MOH- MOE letters of support and addition of new head of NCD-MH & I Department at MoH Outline proposal approved Aug-Sep 2017 Calls and informal meetings MoH and SHN leads 12th Sept 2017 Revised Letter of Support from MoH and MoE through SHN 20th Sept 2017 Full Grant application Submission 1st November 2017 Full proposal approved ACTIVITY ON GRANT 11th June 2018 Approval Letter 8th November 2018 MoH-MoE Stakeholder meeting- Discussed name of trial, need to decide focal ministry, iodine reduction, other stakeholders to include 7th December 2018 MoE Stakeholder meeting at MoE offices- to rope in relevant departments and affiliates- 8th Feb 2019 MoH, MoE Stakeholder meeting- included other stakeholders (university, NGOs and business entities) April 2019 1) Department of Nutrition and HIV/AIDS (DNHA), NCD- MH & I and SHN (MoH)- discussion on lead department- 2) District Commission (Lilongwe and Karonga)- Gave approval to approach District Executive Committee (DEC), District Education Office and other relevant stakeholders (Primary Education Advisors-PEAs, SHN coordinator) 3) Directorate of Quality Assurance and Standards (DQAS)- Discussed the possibility of adding lessons as part of the curriculum. April- May 2019 Translation of curriculum and drawing of illustrations 02nd May 2019 Presentation to Karonga District Management Team and vote of approval 3rd May 2019 Presentation to Lilongwe City council and vote of approval . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 10 06-08th May 2019 Curriculum development workshop (DNHA, SHN, NCD-MH & I, DQAS, MIE, BASIC) May 2019 Editing and formatting of curriculum to match Malawi primary school curriculum by MIE and approved through DQAS. May- June 2019 Meetings with head teachers, Parent teacher associations (PTAs) and School Management committees (SMC) June 2019 Teacher’s training June-July 2019 Pilot trial 09-11th September 2019 Curriculum refinement workshop (DNHA, SHN, NCD- MH &I, DQAS, MIE, BASIC) 167 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 11 This preliminary work included developing and adapting the salt reduction programme using 168 the content of the School-EduSalt Trial conducted in Changzhi, northern China[6] as a starting 169 point, to suit the needs of Malawian adolescents and fit with the existing primary science and 170 technology education curriculum in which health education is integrated . This work included 171 adapting content, illustrations, and food examples to suit the Malawian conte xt. Prior to 172 adaptation the materials were translated to English, and the elements being shared with the 173 learners families, were translated after adaptation into the local languages (Chichewa/ 174 Chitumbuka). Learning materials were edited and formatted to be consistent with other science 175 topics during a stakeholder workshop with two MoH departments (Department of Nutrition 176 and HIV/AIDS- DNHA & Non- Communicable Diseases, mental health and Injury NCD- MH 177 & I), four MoE departments (School Health Nutrition- SHN, Directorate of Quality Assurance 178 and Standards - DQAS, Basic Education (BASIC) Department and Department of Teacher 179 Education and Development -DTED) and one Ministry of Education affiliate organization 180 (MIE), (Fig 1). 181 182 Phase 2: Process evaluation of a programme pilot, including identification of 183 barriers and facilitators to successful implementation 184 The pilot programme took place in two primary schools in urban Lilongwe and rural Karonga. 185 Standard 6 t eachers received three days of training that introduced the programme, the 186 intervention, and associated resources ( e.g., parent letters, posters) and activities ( e.g., parent 187 forum, quiz and artistic performances). All learners in “Standard 6” (sixth year of universal 188 free primary education in Malawi) received the programme for 12 weeks. The programme 189 included weekly lessons that covered salt and health, harmful effects of salt and guidance on 190 salt reduction strategies. Four lesson summary letters (in local language) were sent home with 191 learners for parents to read. Three of the letters asked the parents to respond to several simple 192 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 12 questions from the summary as an assessment of their understanding of salt intake and 193 hypertension, further engaging parents and indicating the number of actively participating 194 parents. Learners also participated in after -school activities either on their own i n artistic 195 competitions (poems, drama , song performance s) or with a parent during family quizzes. 196 Parents participated in a quiz and forum discussing their experience with the programme. 197 Teachers purposefully selected and invited learners to Focus Group Discussions (FGDs). 198 Children took home letters with an information sheet detailing the trial information and consent 199 forms (in the local language) for parents and guardians, which they returned after being signed. 200 We conducted four FGDs comprising ten learners each who had participated in the pilot (Table 201 2). 202 Table 2 . Summary of phase 2 data collection methods and participant 203 characteristics 204 Participant type Data collection method Number of interviews/ FGDs Number of participants Head teachers Semi-structured interviews 2 2 Teachers Group interviews 2 4 Parents FGD 3 (one FGD males/one females/one mixed gender) 30 Learners FGD 4 (two with boys, two with girls) 40 205 A convenience sample of parents and guardians were invited to participate in FGDs via a letter 206 and information sheet taken home by learners. Teachers were also invited to participate in 207 FGDs which lasted 40-60 minutes. Semi-structured interviews were held with head teacher s 208 from each school. 209 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 13 A semi-structured interview guide was used to gain perspectives on the learners’ views of the 210 programme, including what they liked or disliked and their suggestions for improvement . For 211 parents, we explored their experience s with the programme, the learning from their children 212 and their interaction s with teachers and the research team through parent meetings. We 213 conducted seven non-participatory observations of three lessons delivered to learner s. 214 Interviews and FGDs were audio recorded and transcribed verbatim with the transcripts serving 215 as data for analysis. 216 Data analysis 217 Phase 1 and 2 data (minutes, field notes, transcripts, and observations) were coded in NVivo 218 12 using a thematic approach to analysis as outlined by Braun and Clarke.[19] Transcripts were 219 initially coded by three separate researchers to develop a codebook. These researchers 220 discussed any disagreements in the coding through virtual meetings fortnightly to ensure 221 consistency of coding. Once the coding frame was established, all data was double coded. Field 222 notes, minutes and transcripts were analysed using a thematic approach to identify participants’ 223 main concerns on the salt reduction education and their perspectives on the curriculum, 224 delivery mechanisms, family and community involvement as well as policy implications. The 225 identified them es were then mapped onto the four main constructs of the NPT theore tical 226 framework; coherence (sense-making), cognitive participation (engagement work), collective 227 action (operationalization work), and reflexive monitoring (appraisal) [18] to help 228 conceptualize the data. Themes that f ell outside the framework were noted but not excluded 229 from analysis. In this way, we avoid “shoe-horning” of the data and help identify issues that 230 might still be important but not fit with our theoretical mode l. Finally, a descriptive summary 231 was written to capture the information (with exemplar quotes) in each broad theme. The notes 232 from the formative work were summarized to form a description of the processes to inform 233 implementation of the programme during the trial. 234 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 14

235 Phase 1: Formative work 236 Issues addressed and resolved during stakeholder meetings are as outlined below. 237 1. Leadership and departmental involvement 238 A key question was which Ministry would be considered the focal or secretariat Ministry for 239 the trial. As a health-related trial embedded in schools, clarification was required whether it 240 should be under Ministry of Health or M inistry of Education (Fig 1) and positively, both 241 Ministries were keen to assume leadership . Nutrition related initiatives in Malawi are often 242 spearheaded by the Department of Nutrition and HIV/AIDS (under the Ministry of Health ). 243 However, this trial was deemed unique because it involved delivery of the intervention as part 244 of the science curriculum, hence the Ministry of Education felt the trial should be overseen by 245 their Ministry. Even within the M inistry of Health, there was uncertainty whether the trial 246 should be overseen by the department of Non-Communicable Diseases and Mental Health & 247 Injuries (NCD-MH & I) or the Department of Nutrition and HIV/AIDS (DNHA). After internal 248 discussions the Ministry of Health assumed the lead through the DNHA. Department of School 249 Health and Nutrition (SHN) was mandated by DNHA to oversee trial implementation because 250 of its links to both the Ministry of Health and Ministry of Education (Table 1 and Fig 1). School 251 Health and Nutrition with NCD- MH & I worked with the study team during implementation 252 of the study while DNHA was to be engaged at a minimum capacity , with a later role of 253 translating results into nutrition policy. Despite reaching agreement there were residual 254 concerns from other relevant government departments and SHN was instrumental in achieving 255 final consensus and encouraging full participation. 256 Further smaller meetings provided an opportunity for the School Health and Nutrition 257 Department to identify other key departments within the Ministry of Health and Education to 258 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 15 be included in the next phase of meetings e.g. Department of Quality Assurance and Standards 259 (DQAS) responsible for supervision and inspection of teaching and learning resources in 260 schools, Basic Education (BASIC) responsible for teacher’s welfare , Health Education Unit 261 (HEU) responsible for promoting health education and Malawi Institute of Education (MIE) 262 an affiliate of the Ministry of Education mandated to design, develop, monitor and evaluate 263 primary school and teachers curriculum. 264 2. Impact on iodine fortification programme 265 There was a concern from the DNHA representatives, that the intervention might affect iodine 266 intake, as salt in Malawi is fortified with iodine as a public health measure to prevent goitre. 267 Presentations by the study team were made using Malawi data from the Public Health Institute 268 demonstrating that children and adults have a consistent excess iodine intake in all districts of 269 Malawi and highlighting the tension between salt advertising capitalizing on the iodization 270 programme (e.g., “Salt is good for your health”) and picturing high levels of salt application to 271 food. 272 Another concern was the study title: ‘No to Na’ that might suggest total cessation of salt intake 273 and lead to inadequate iodine intake undermining the extensive work that had previously been 274 undertaken to improve iodine intake over the years. The NCD- MH & I department highlighted 275 that health care workers are already advocating for reduction of salt as management or 276 treatment strategy for hypertension and other NCDs. It was agreed to use an alternative title 277 during the trial, and in interactions at district and community level. The study title was “Healthy 278 Diets Study” to reduce risk of disclosure of the intention of the behavioural intervention in the 279 control schools while maintaining “No to Na ” as the protocol title as submitted to Ethics 280 Committee. 281 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 16 3. Integration into science curriculum 282 The Ministry of Education through the Department of Quality Assurance and Standards 283 (DQAS) expressed concerns regarding the propos al to integrate the intervention into the 284 science curriculum. The DQAS was concerned that with the school curriculum for the year 285 already set, the only option was to offer it as standalone lessons. DQAS reviewed the proposed 286 lessons with the support of the Malawi Institute of Education (MIE) and approved them to be 287 piloted as standalone lessons, one lesson at the end of the day , during standard school hours, 288 once every week. 289 After resolving these issues with key ministry stakeholders, a kick-off meeting was held at the 290 beginning of 2019 with fifteen attendees (Table 1). The meeting was held to introduce and 291 discuss the trial with a larger and diverse stakeholder group. 292 Subsequent meetings were held at the district level of government with the district executive 293 committee (DEC) in Lilongwe (32 attendees) and Karonga (53 attendees) , technical 294 committees that advises and support s the District Councils (Table 1). As the engagement 295 continued other smaller office meetings were held with people directly involved in supervising 296 the schools e.g. the District Education Office, Primary Education Advisors (PEA) and School 297 Health and Nutrition coordinators. 298 4. Resources required 299 Although extensive stakeholder engagement was crucial to ensure successful implementation 300 of the trial overall, the trial team had not anticipated the number and size of meetings that would 301 ultimately be required. Pre-application discussions had not identified the number of different 302 Ministry departments that would need to be involved, nor the complexity of processes required 303 to generate and authorise the materials to be used . This important formative work resulted in 304 increased study costs to cover conference, subsistence, travel and accommodation costs and 305 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 17 delayed the start of piloting. Stakeholder engagement began in September 2018 and the pilot 306 took place in June-July 2019. 307 5. Development of teaching materials 308 Before the pilot beg an, materials from the Edu -Salt programme[6, 15] were translated from 309 Mandarin to English to generate a template. The materials were then significantly adapted to 310 reflect the Malawian context, taking due note of the major difference between urban and rural 311 diets and lifestyles. The study team originally proposed separate materials for urban and rural 312 communities, but the Ministry of Education (MoE) was clear that if successful, in order to 313 make the programme scalable only one set of materials should be used . The study team also 314 proposed that all materials be in the local languages (Chichewa for urban Lilongwe, and 315 Chitumbuka for rural Karonga). The MoE team were again clear that English was the official 316 language of instruction for Science in Standard 6 and that lessons must be conducted in English 317 with English materials. Concerns about this were addressed by the Education team who were 318 clear that that the teachers may and do explain difficult words or extracts to learners in their 319 local language to help them gr asp the topic, then switch back to English . The exception that 320 was negotiated was the materials to be shared with the parents ( letters and flyer) which were 321 to be prepared in the two relevant languages. A local illustrator was engaged to add illustrations 322 to match the content in the curriculum (Fig 2). 323 The materials and programme were discussed in detail at a programme development workshop 324 that included the key Ministry departments and affiliates as well as teachers and head teachers 325 (21 attendees). Following this workshop, the programme was further modified and tailored to 326 fit the Malawian curriculum format. 327 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 18 328 Fig 2. Flyer for parents showing strategies for reducing salt 329 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 19 Phase 2: process evaluation of the programme pilot 330 The findings from the pilot of the programme are presented under five key themes: 1) Sense-331 making: teacher, parent and learner views of a school -based intervention for reducing salt 332 (coherence); 2) Relationship wo rk: engaging learners and through them parents in the salt 333 reduction programme (cognitive participation) ; 3) Enacting work : teachers, parents and 334 learners doing the work of participating in the programme (collective action); 4) Perceived 335 facilitators to the programme (reflexive monitoring); 5) Perceived challenges to the programme 336 (reflexive monitoring). These themes are described, and illustrative quotations provided. 337 1. Sense-making: teacher, parent, and learner views of a school -based 338 programme for reducing salt intake (coherence). 339 Learners and parents were positive about the programme and were able to identify possible 340 adverse effects of a high -salt diet. Participants also recognised their households’ high salt 341 intakes and spoke about how they were able to reduce this and become accustomed to the taste 342 of less salty food. The learners’ FGDs demonstrated learning relating to each of the programme 343 units including the harmful effects of excessive salt, though some had difficulties 344 understanding the recommended daily salt intake and how best to measure and reduce their salt 345 intake. Some parents also spoke of similar challenges as highlighted below: 346 The way I understood from the several meetings I have been. This is the second one. I 347 heard that the whole household is supposed to eat less than 5 grams per day. So, you 348 will correct me if am wrong right here [Interviewer 2: we need to correct each other] 349 we should eat less than 5 grams of salt [I nterviewer 1: Per person?] no per family. 350 [Parent] 351 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 20 Others spoke of putting their learning into practice; and some interviewees linked this new 352 learning to their existing knowledge, for example about people in their family affected by 353 relevant conditions such as high blood pressure as shown below: 354 “I: So why did you decide to tell your parents? 355 P: To protect their life as my mother already has ulcers and my father has B lood 356 Pressure, so we were taught that even our parents should not eat too much salt.” 357 [Learners] 358 2. Relationship work: engaging learners and through them parents in the 359 salt reduction programme (cognitive participation) 360 Learners told their extended families and neighbours about what they had learn t about salt 361 reduction. This communication was supported by the letters from schools to parents. 362 When the child got home, he said “I have come with this letter from KPS [Karonga 363 Prevention study now known as Malawi Epidemiology and Intervention Research Unit-364 MEIRU] who were teaching us at school” then I asked hi m, what did they teach you? 365 He said, they gave us a letter and told us to give to our father, so here is the letter. I got 366 the letter as an individual and started reading. When I finished the letter, I invited my 367 family and at home I have a mother, my sist er, two wives and children, then I started 368 explaining. [Parents] 369 Parents were invited to a parent forum and then a quiz. At the parent forum, parents discussed 370 the work their children did in class and at the quiz, two groups of parents competed with one 371 another, answering questions about salt reduction. Teachers thought engaging parents 372 consolidated the information that learners delivered at ho me and meant parents could work 373 through challenges to reducing salt intake. Parents valued these meetings as it allowed them to 374 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 21 learn from other parents and these meetings motivated parents and learners to participate in the 375 intervention as illustrated below. 376 “When we took part in the quiz, we were encouraged that we are not alone in this study 377 seeing our fellow parents participating and even the children were encouraged seeing 378 parents there, thank you very much.” [Parent] 379 Some parents said they agreed to participate because of previous experiences with the research 380 group, MEIRU which has shown them that their work is honest and aims to help people. 381 The parents who attended FGDs were generally very positive about the parent meetings, 382 however only half the parents who were invited attended. Some parents explained that the non-383 attenders could not come as they worked or went to church on Sundays, but others maintained 384 that they were negligent. Most parents said that providing incentives, such as drinks or snacks, 385 would encourage more parents to attend meetings. Teachers also valued incentives as this quote 386 shows. 387 And we felt when you were coming to our schools to do that exercise with learners, 388 after classes we thought the learners will be given a little something whether Mahewu 389 [maize drink] just to keep them up to the knocking time. [Head teacher] 390 Parents’ sugge stions for improving participation included: a) providing incentives/gifts to 391 encourage parent attendance; b) sending meeting invitations well in advance to give parents 392 time to prepare; c) sharing meeting minutes with absentee parents through their children so that 393 they could see what they missed; d) engaging with the community through the village chiefs, 394 where the intervention could be explained to everyone. 395 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 22 3. Enacting work: teachers, parents and learners doing the work of 396 participating in the programme (collective action) 397 Learners, parents, and teachers described the practical steps they took to reduce salt. 398 Participants explained how they reduced their salt intake, some telling stories of, for example, 399 planting garlic to provide an alternative flavouring o r buying a spoon for measuring the salt. 400 Others said they had stopped putting salt on the table or hid it from others in their house to 401 ensure that they did not add extra salt to food. 402 “I have already explained about challenges which I face. My other fami ly members 403 complain whenever I add little salt and they add extra salt. So, I take that salt and hide 404 it somewhere so that they should think that we have run out of salt.” [Parent] 405 This behaviour change met challenges. Parents had difficulties in reducing salt intake as some 406 family members (or visitors) resisted, as they found it difficult to become accustomed to the 407 new taste as this parent explains. 408 “…the challenge which I faced was we were separating relish because my husband was 409 refusing to take low salt. He was saying he can’t manage to follow this advice because 410 he is accustomed to taking more salt and he can’t start taking little salt. He also said, 411 this is unhealthy because a person can’t eat relish which he/she feels does not taste 412 good.” [Parent] 413 Learners also reported some resistance to reducing salt intake from relatives. Some learners 414 talked about resistance from their parents who did not want to listen to their children and who 415 did not want to reduce their salt intake. This learner explains: 416 “…when I am telling them they say, “it’s just a topic which you are just learning at 417 school so that your future should be bright” so it’s what they say.” [Learner] 418 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 23 Some parents agreed that in the beginning it was difficult for them to listen to what their 419 children were saying (though they did eventually take up the intervention). 420 4. Perceived facilitators to the programme (reflexive monitoring) 421 There was positive feedback from learners, parents, and teachers about the programme and 422 they reported the positive impact that the programme had in their lives. Some felt that their 423 reduced salt intake had significantly improved their health, as this learner explains: 424 “I was one of the people who liked to add salt to foods that was already salted whether 425 enough or a lot, I just had to add so I can taste saltiness. But after I have learnt I know 426 that it’s good not to add salt to food that is already salted. It has helped my life. ” 427 [Learner] 428 Teachers felt the programme was beneficial to the nation as the information gained by learners 429 would spread to parents, neighbours, and the community. A teacher explains: 430 “…the project was a good one in a sense that we have learnt a lot from it. One, how to 431 control our diet, how to control our sugar levels, how to control our salt levels and how 432 we can manage ourselves. And most of all, learners have learnt a lot from it, in a sense 433 that they have benefited something from it… So, we feel the project has a great impact 434 to the nation because once a learner is taught, he will still be imparting that knowledge 435 to the community, to the family members and to their friends as well. [Teacher] 436 According to the parents’ FGDs, potential improvements to their lives, and crucially to their 437 children’s lives, was the greatest motivation for engaging with the programme. Parents spoke 438 of how their children were the future, and that the children liked to see their parents 439 participating in meetings at the school. 440 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 24 Parents and teachers had suggestions about future expansion of the programme. They wanted 441 the programme to be incorporated into the syllabus and expanded to the whole of Malawi but 442 also to include other classes , as it was a program me which could benefit ma ny learners and 443 families. 444 “I just want to say you should continue with the program because a lot of people don’t 445 know about salt, so reach out even in villages reach out to them, some don’t know that 446 too much salt intake is harmful, even here in town, some people don’t know about salt.” 447 [Parent] 448 5. Appraisal work: perceived challenges to the intervention (reflexive 449 monitoring) 450 While teachers felt the programme was valuable, they noted some challenges, for example, 451 the terminology used in the learning materials was unfamiliar to learners and made lessons 452 difficult. Teachers and learners also mentioned language as a challenge as the learners had 453 just transitioned from using Chichewa as a language of instruction to English in Standard 5. 454 “…of course, in class we have active learners, and we also have learners that learn at 455 a slower pace, so we tried so that these slower pace learners should understand because 456 our aim was the learners must teach the parents in the community and we could even 457 try to translate the content from English to their local language for them to understand 458 and take the message to their parent. [Teacher] 459 Some of the challenges that teachers faced were a lack of time to complete the programme 460 (some lessons had to be held outside normal school hours) and a lack of teaching materials (or 461 unsuitable materials). 462 “…the only problem we had was on time; th e time was really short, and I can say the 463 content was over planned, but we tried our best to take time to teach and mainly the 464 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 25 last period we used it to extend the period with 30 minutes for us to finish the lesson.” 465 [Teacher] 466 Individual schools and teachers also needed to supply some materials, such as ingredients to 467 cook porridge (for a class experiment) , this was difficult as the programme started midway 468 through the school year when budgets had already been allocated but the ministry of Education 469 had made it clear that teaching aids are prepared by either the school or the teacher. 470 “…the challenges that we met, mainly we had inadequate teaching and learning 471 resources. Like we had to have packets of different types of salt. We are talking of 472 kitchen salt, table salt which we can distribute in groups so that every group should 473 have a look at the packet of salt which is not open. So, to get those, it was really a 474 problem. [Teacher] 475

476 This paper presents: 1) a description of extensive formative work undertaken with stakeholders 477 to develop and refine a school-based salt reduction programme in preparation for a cluster 478 randomised trial; and 2) the findings of the process evaluation undertaken during a pilot of the 479 programme. 480 Stakeholder engagement preceding trial implementation in a LMIC is rarely reported and based 481 on the findings presented here it is clear that it is time consuming and requires substantial 482 investment. However, this effort was essential to obtain the support of key stakeholders, avoid 483 misconceptions and allay any concerns, as well as to clarify which stakeholders would be 484 responsible for implementation of the programme . This work reaped benefits as outlined in 485 Table 3 and facilitated approvals for trial activities and minimised potential delays. 486 487 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 26 Table 3– Summary of key achievements and implications of the formative work 488 Key achievements for formative work Impact Agreed on leadership and key contacts for government departments and their roles in the deployment of the salt reduction programme in schools. 1) Provided a sense of ownership and helped achieve “buy in” of key stakeholders. 2) Improved flow of information, minimised approval delays and resolved conflicts. Addressed concerns about the trial with key government ministries and departments (Fig 1) and reached a shared understanding of the aims and

of the trial. Removed barriers to deployment of the intervention (for example, by alleviating concerns that the intervention might reduce iodine consumption or stop salt consumption completely and undermine previous public health messaging.) Enabled identification and inclusion of relevant and diverse group of stakeholders. The diversity of input optimised development and refinement of the intervention materials. 489 The process evaluation of the programme pilot then identified aspects of the programme that 490 needed to be refined or altered, for example, in relation to timing/duration of the lessons , 491 content of the lessons and teaching resources, to improve the likelihood of successful 492 implementation of the programme (see Table 4). Delivery of the programme was facilitated 493 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 27 by the ease with which it could be implemented with minimal disruption to an already packed 494 curriculum. 495 Table 4 – Summary of challenges identified in the programme pilot and implications for 496 the trial. 497 Challenge Key changes Lessons took longer than the allotted time – requiring teachers to work overtime and children to stay later at school Some lessons were split into two separate lessons to enable the teaching to be incorporated within the school day. Holding children longer outside school hours Lesson was shifted from end of the day to mid-morning Content of lessons above learner proficiency Content wa s further simplified, and a glossary was provided with meanings Language was a barrier to learners fully understanding the intervention as children at this point in their education were reasonably new to science lessons taught in English Teachers had the flexibility to explain difficult concepts in the local language Resources too limited to accommodate all the learners. Produced extra posters and each learner was given a book of the curriculum. 498 499 The process evaluation of the programme pilot showed that parents, children, and teachers were 500 generally positive about the programme. Students had the opportunity to contribute to the 501 learning process by sharing their views on the topics. 502 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 28 Previous research has suggested that prevention programmes are most effective if the approach 503 is multilevel, involving the learner, school and family .[20] Similarly, a report published by 504 WHO,[21] concluded that the most effective school interventions are multicomponent and 505 should include a curriculum taught by trained teachers, supportive school policies, a physical 506 activity programme and healthy food served by the school canteen services (which rarely exist 507 in the local setting). In this salt reduction programme participation in parent forums or quizzes 508 was suboptimal and participants believed this was due to lack o f incentives ( e.g. snacks, 509 money), which although they could have been provided in a trial setting would have made the 510 intervention unsustainable. However, parents enjoyed parent forums as it showed them that 511 they were not alone in their experiences with the programme and they believed their 512 participation encouraged their children to comply with the programme. 513 Over the past two decades, school-based health programmes have become a key component of 514 public health strategies in LMICs. For example, they have been used in the Middle East, Europe, 515 Asia, Africa, and South America to promote adolescent mental health ,[22] and in Zambia, 516 South Africa, Tanzania and Belize to promote AIDS awareness and sexual and reproductive 517 health.[23] 518 School-based health programmes focusing on diet and nutrition have included standards placed 519 on the composition of school meals and limits to food choices available in school canteens as 520 well as changes to school curricula to improve nutrition education.[21] School-based nutrition 521 and diet interventions have been widely used with a recent umbrella review identifying thirteen 522 separate systematic reviews describing eighty-two studies undertaken in the United States, 523 Mexico, Canada, Europe, Asia, South America, the Middle East and Australia.[24] 524 A review of childhood obesity prevention interventions in Africa,[25] included 17 articles 525 describing 14 interventions in three countries (South Africa, Tunisia and Uganda). This review 526 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 29 identified barriers to implementing school -based interventions as largely resource -related, 527 including lack of time from teachers and stakeholders, buy -in, training or motivation. Some 528 interventions in China[26-28] (which were not focused on salt) involved parents and Wang et 529 al.’s[29] study of a school -based initiative to reduce obesity noted that interactive and 530 innovative intervention components facilitated the intervention. 531 Apart from No to Na, the only other interventions focused solely on salt reduction were the 532 Edu-salt programme in China,[6] on which the No to Na intervention is based and the COMBI-533 ELS intervention in Vietman.[30] The COMBI-ELS intervention involved several components 534 including targeted primary school interventions , mass media communications, training for 535 cooks, and community programmes. I t was not possible to ascertain which elements of the 536 programme were most effective in changing behaviour nor was it possible to assess the 537 contribution of the work in schools to the programme’s success. 538 A recent systematic review [31] examining the effectiveness of school -based nutrition 539 interventions in sub-Saharan Africa found 14 studies based in South Africa, Botswana, Burkina 540 Faso, Kenya, Nigeria, and Tanzania, but none focused on salt reduction, instead they focused 541 on improving diets to reduce obesity and increasing physical activity. These studies showed 542 that there was an improvement in nutrition knowledge following the interventions, but this did 543 not necessarily translate into healthy nutrition behaviour. Just one school based intervention in 544 South America (Trinidad and Tobago) included salt-reduction in its focus (in this case reducing 545 the consumption of snack foods high in fat, sugar and salt) as part of a wider programme.[32] 546 Strengths and limitations 547 A key strength of this study is that it describes the form ative work undertaken to develop the 548 programme and gives an account of the extensive engagement work required. The authors have 549 not found any other reports of such work, though it is essential to inform future research and 550 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 30 policy. Rigorous thematic analysis was conducted, drawing on both inductive (i.e., data-driven) 551 and deductive (i.e., based on pre-conceived ideas) approaches, until data saturation was reached, 552 and our analysis was underpinned by a robust theoretical framework, N ormalization Process 553 Theory. Double coding of all pilot process evaluation interviews and coding clinics enhanced 554 reliability. However, the work was undertaken in Malawi and the findings cannot necessarily 555 be generalised to other LMICs. We did not interview those who did not engage with the 556 programme and so respondents may provide a more positive view of the intervention than their 557 peers who did not engage. 558

559 This study enhance s our understanding of how a range of health, nutrition and education 560 stakeholders view, experience, and value a school-based programme for salt reduction, and of 561 the factors that may influence implementation and impact of the programme in a LMIC. Our 562 work has highlighted the importance of extensive, preliminary engagement with a wide range 563 of stakeholders to achieve “buy in” and active cooperation but has also demonstrated that future 564 studies of this kind need to allocate substantial time and funds to this task. Our process 565 evaluation identified five key challenges that led to refinements to how the programme was 566 delivered. Our parallel process evaluation of the programme as planned to be implemented 567 during the trial will help us understand whether these refinements helped and what further 568 changes, if any, would be required to facilitate scale up beyond the trial or implementation in 569 similar settings. Work such as this, which generates insights about the feasibility, acceptability, 570 usability, and up -scalability of school-based programmes, is crucial for practice and policy 571 decision-making on the future provision of novel health programmes in school settings. 572 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 31

573 We would like to thank all the participants who participated in the intervention and in the focus 574 group discussions o r interviews. We owe the head teachers, teachers, learners, and their 575 guardians from the two primary schools our gratitude for their time, patience, and contributions. 576 We also acknowledge and thank all key stakeholders who participated and contributed to t he 577 formative work. We would like to thank Hazel Namadingo and Cecilia Nyirenda for all their 578 work and effort in this study. Gratitude should also go to James Kazembe for the tremendous 579 work in the illustrations used in the intervention materials. 580 Disclosure statement 581 No conflict of interest 582 Funding 583 This work was supported by the Medical Research Council, DfiD/NIHR United Kingdom 584 through the Adolescent Health in low-income countries scheme [MR/R022186/1] 585

586 1. Prof Rafael Lozano MN. Global and regional mortality from 235 causes of death for 20 age 587 groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. The 588 lancet. 2012;380(9859):2095-128. doi: https://doi.org/10.1016/S0140-6736(12)61728-0. 589 2. Price AJ, Crampin AC, Amberbir A, Kayuni-Chihana N, Musicha C, Tafatatha T, et al. 590 Prevalence of obesity, hypertension, and diabetes, and cascade of care in sub-Saharan Africa: a 591 cross-sectional, population-based study in rural and urban Malawi. The lancet Diabetes & 592 endocrinology. 2018;6(3):208-22. Epub 2018/01/27. doi: 10.1016/s2213-8587(17)30432-1. PubMed 593 PMID: 29371076; PubMed Central PMCID: PMCPMC5835666. 594 3. Organization WH. Noncommunicable diseases progress monitor. Switzeland: World Health 595 Organization, diseases DfpoN; 2017. 596 4. Niinikoski H, Jula A, Viikari J, Rönnemaa T, Heino P, Lagström H, et al. Blood pressure is lower 597 in children and adolescents with a low-saturated-fat diet since infancy: the special turku coronary 598 risk factor intervention project. Hypertension (Dallas, Tex : 1979). 2009;53(6):918-24. Epub 599 2009/04/15. doi: 10.1161/hypertensionaha.109.130146. PubMed PMID: 19364991. 600 5. Muthuri SK, Oti SO, Lilford RJ, Oyebode O. Salt Reduction Interventions in Sub-Saharan 601 Africa: A Systematic Review. PloS one. 2016;11(3):e0149680. Epub 2016/03/11. doi: 602 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 32 10.1371/journal.pone.0149680. PubMed PMID: 26963805; PubMed Central PMCID: 603 PMCPMC4786148. 604 6. He FJ, Wu Y, Feng XX, Ma J, Ma Y, Wang H, et al. School based education programme to 605 reduce salt intake in children and their families (School-EduSalt): cluster randomised controlled trial. 606 BMJ (Clinical research ed). 2015;350:h770. Epub 2015/03/20. doi: 10.1136/bmj.h770. PubMed 607 PMID: 25788018. 608 7. Ha SK. Dietary salt intake and hypertension. Electrolyte & blood pressure : E & BP. 609 2014;12(1):7-18. Epub 2014/07/26. doi: 10.5049/ebp.2014.12.1.7. PubMed PMID: 25061468; 610 PubMed Central PMCID: PMCPMC4105387. 611 8. Organization WH. Guideline: sodium for adults. Switzerland: World Health Organization, 612 2012 2012. Report No. 613 9. Mozaffarian D, Fahimi S, Singh GM, Micha R, Khatibzadeh S, Engell RE, et al. Global sodium 614 consumption and death from cardiovascular causes. The New England journal of medicine. 615 2014;371(7):624-34. Epub 2014/08/15. doi: 10.1056/NEJMoa1304127. PubMed PMID: 25119608. 616 10. Noubiap JJ, Essouma M, Bigna JJ, Jingi AM, Aminde LN, Nansseu JR. Prevalence of elevated 617 blood pressure in children and adolescents in Africa: a systematic review and meta-analysis. The 618 Lancet Public health. 2017;2(8):e375-e86. Epub 2017/12/19. doi: 10.1016/s2468-2667(17)30123-8. 619 PubMed PMID: 29253478. 620 11. Ataklte F, Erqou S, Kaptoge S, Taye B, Echouffo-Tcheugui JB, Kengne AP. Burden of 621 undiagnosed hypertension in sub-saharan Africa: a systematic review and meta-analysis. 622 Hypertension (Dallas, Tex : 1979). 2015;65(2):291-8. Epub 2014/11/12. doi: 623 10.1161/hypertensionaha.114.04394. PubMed PMID: 25385758. 624 12. Simons-Morton DG, Obarzanek E. Diet and blood pressure in children and adolescents. 625 Pediatric nephrology (Berlin, Germany). 1997;11(2):244-9. Epub 1997/04/01. doi: 626 10.1007/s004670050271. PubMed PMID: 9090675. 627 13. Oyebode O, Oti S, Chen YF, Lilford RJ. Salt intakes in sub-Saharan Africa: a systematic review 628 and meta-regression. Population health metrics. 2016;14:1. Epub 2016/01/14. doi: 10.1186/s12963-629 015-0068-7. PubMed PMID: 26759530; PubMed Central PMCID: PMCPMC4709973. 630 14. Bazzano LA, Green T, Harrison TN, Reynolds K. Dietary approaches to prevent hypertension. 631 Current hypertension reports. 2013;15(6):694-702. Epub 2013/10/05. doi: 10.1007/s11906-013-632 0390-z. PubMed PMID: 24091874; PubMed Central PMCID: PMCPMC4366416. 633 15. He FJ, Wu Y, Ma J, Feng X, Wang H, Zhang J, et al. A school-based education programme to 634 reduce salt intake in children and their families (School-EduSalt): protocol of a cluster randomised 635 controlled trial. BMJ open. 2013;3(7). Epub 2013/07/19. doi: 10.1136/bmjopen-2013-003388. 636 PubMed PMID: 23864214; PubMed Central PMCID: PMCPMC3717470. 637 16. Organization WH. Global action plan for the prevention and control of noncommunicable 638 diseases: 2013-2020.2013 2013. 639 17. May CR, Cummings A, Girling M, Bracher M, Mair FS, May CM, et al. Using Normalization 640 Process Theory in feasibility studies and process evaluations of complex healthcare interventions: a 641 systematic review. Implementation science : IS. 2018;13(1):80. Epub 2018/06/09. doi: 642 10.1186/s13012-018-0758-1. PubMed PMID: 29879986; PubMed Central PMCID: PMCPMC5992634. 643 18. Murray E, Treweek S, Pope C, MacFarlane A, Ballini L, Dowrick C, et al. Normalisation process 644 theory: a framework for developing, evaluating and implementing complex interventions. BMC 645 medicine. 2010;8(1):63. 646 19. Braun V, Clarke V. What can "thematic analysis" offer health and wellbeing researchers? 647 International journal of qualitative studies on health and well-being. 2014;9:26152. Epub 648 2014/10/19. doi: 10.3402/qhw.v9.26152. PubMed PMID: 25326092; PubMed Central PMCID: 649 PMCPMC4201665. 650 20. Bindler RC, Goetz S, Butkus SN, Power TG, Ullrich-French S, Steele M. The process of 651 curriculum development and implementation for an adolescent health project in middle schools. The 652 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint 33 Journal of school nursing : the official publication of the National Association of School Nurses. 653 2012;28(1):13-23. Epub 2011/10/14. doi: 10.1177/1059840511424414. PubMed PMID: 21994321. 654 21. Organization WH. Creating an enabling environment for population-based salt reduction 655 strategies: report of a joint technical meeting held by WHO and the Food Standards Agency, United 656 Kingdom, July 2010. 2010. 657 22. Fazel M, Patel V, Thomas S, Tol W. Mental health interventions in schools in low-income and 658 middle-income countries. The Lancet Psychiatry. 2014;1(5):388-98. doi: 10.1016/S2215-659 0366(14)70357-8. 660 23. Mason-Jones AJ, Sinclair D Fau - Mathews C, Mathews C Fau - Kagee A, Kagee A Fau - 661 Hillman A, Hillman A Fau - Lombard C, Lombard C. School-based interventions for preventing HIV, 662 sexually transmitted infections, and pregnancy in adolescents. Cochrane Database Systematic 663 Reviews. 2016;(1469-493X (Electronic)). 664 24. O'Brien KA-O, Barnes C, Yoong SA-OX, Campbell E, Wyse R, Delaney TA-O, et al. School-665 Based Nutrition Interventions in Children Aged 6 to 18 Years: An Umbrella Review of Systematic 666 Reviews. LID - 10.3390/nu13114113 [doi] LID - 4113. Nutrients. 2021;(2072-6643 (Electronic)). 667 25. Shieh ANBaJ. School-based physical activity and diet interventions can be effective for 668 addressing obesity in children in LMICs: Evidence from systematic reviews 2020 [updated 2020-11-669 13T08:48:02+00:00]. Available from: https://researchforevidence.fhi360.org/school-based-physical-670 activity-and-diet-interventions-can-be-effective-for-addressing-obesity-in-children-in-lmics-671 evidence-from-systematic-reviews. 672 26. Cao Z-J, Wang S-M, Chen Y. A randomized trial of multiple interventions for childhood 673 obesity in China. American journal of preventive medicine. 2015;48(5):552-60. 674 27. Meng L, Xu H, Liu A, van Raaij J, Bemelmans W, Hu X, et al. The costs and cost-effectiveness 675 of a school-based comprehensive intervention study on childhood obesity in China. PloS one. 676 2013;8(10):e77971. 677 28. Zhou W-j, Xu X-l, Li G, Sharma M, Qie Y-L, Zhao Y. Effectiveness of a school-based nutrition 678 and food safety education program among primary and junior high school students in Chongqing, 679 China. Global health promotion. 2016;23(1):37-49. 680 29. Wang D, Stewart D, Chang C, Shi Y. Effect of a school-based nutrition education program on 681 adolescents’ nutrition-related knowledge, attitudes and behaviour in rural areas of China. 682 Environmental Health and Preventive Medicine. 2015;20(4):271-8. doi: 10.1007/s12199-015-0456-4. 683 30. Do HTP, Santos JA, Trieu K, Petersen K, Le MB, Lai DT, et al. Effectiveness of a 684 Communication for Behavioral Impact (COMBI) intervention to reduce salt intake in a Vietnamese 685 Province based on estimations from spot urine samples. The Journal of Clinical Hypertension. 686 2016;18(11):1135-42. 687 31. Kyere P, Veerman JL, Lee P, Stewart DE. Effectiveness of school-based nutrition interventions 688 in sub-Saharan Africa: a systematic review. Public Health Nutrition. 2020;23(14):2626-36. 689 32. Francis M, Nichols SS, Dalrymple N. The effects of a school-based intervention programme 690 on dietary intakes and physical activity among primary-school children in Trinidad and Tobago. 691 Public health nutrition. 2010;13(5):738-47. 692 693 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2022. ; https://doi.org/10.1101/2022.07.20.22277598doi: medRxiv preprint