RESEARCH ARTICLE Open Access

Stakeholder perspectives on theeffectiveness of the Victorian SaltReduction Partnership: a qualitative studyEmalie Rosewarne1*, Wai-Kwan Chislett1, Briar McKenzie1, Jenny Reimers2, Kellie-Ann Jolly3, Kirstan Corben2,Kathy Trieu1 and Jacqui Webster1

Abstract

Background: Interventions to reduce population salt intake are feasible and cost-effective. The Victorian SaltReduction Partnership implemented a complex, multi-faceted salt reduction intervention between 2014 and 2020 inthe Australian state of Victoria. This study aimed to understand stakeholder perspectives on the effectiveness of theVictorian Salt Reduction Partnership.

Methods: Semi-structured interviews were conducted with Partnership and food industry stakeholders. TheConsolidated Framework for Implementation Research was adapted for the Partnership intervention and used toguide the qualitative analysis.

Results: Fourteen Partnership and seven food industry stakeholders were interviewed. The Partnership was viewedas essential for intervention planning and decision-making and an enabler for intervention delivery. The goals ofcapacity building and collaborative action were perceived to have been achieved. The implementation teamexecuted intended intervention activities and outputs, with some adaptations to strategy. Barriers and enablers toimplementation were identified by interviewees, such as compatibility of individual, organisational and Partnershipvalues and building positive relationships between the Partnership and food industry, respectively. Legal, political,social, environmental, technological and economic factors affecting intervention design, delivery and outcomeswere identified.

Conclusions: Establishing a Partnership with diverse skills and experience facilitated collaborative action, capacitybuilding and execution of the intervention. Monitoring and evaluating implementation informed strategyadaptations, which allowed optimisation of Partnership strategy. The importance of developing strongcommunication networks between strategic and implementation-levels was a key lesson.

Keywords: Salt reduction, Public health nutrition, Stakeholder perspectives, Population intervention

© The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you giveappropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate ifchanges were made. The images or other third party material in this article are included in the article's Creative Commonslicence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commonslicence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtainpermission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to thedata made available in this article, unless otherwise stated in a credit line to the data.

* Correspondence: erosewarne@georgeinstitute.org.au1The George Institute for Global Health, The University of New South Wales,Sydney, NSW 2006, AustraliaFull list of author information is available at the end of the article

Rosewarne et al. BMC Nutrition (2021) 7:12 https://doi.org/10.1186/s40795-021-00414-6

BackgroundSalt reduction interventions have been identified as feas-ible, cost-effective approaches to reduce the non-communicable disease burden attributable to excess saltconsumption [1, 2]. Global estimates suggest salt intakeis double the recommended daily maximum amount of5 g per day [2, 3]. In 2017, diets high in salt resulted inalmost 3.2 million deaths and more than 70 million dis-ability adjusted life years (DALYs) globally [4]. High saltintakes cause high blood pressure, which was the attrib-utable factor in more than 10 million global deaths and381 million DALYs [4]. In an effort to reduce the salt-related non-communicable disease burden, in 2013,United Nations Member States committed to the globaltarget of a 30% relative reduction in average populationsalt intake by 2025 [5].Despite this commitment, coordinated efforts to re-

duce salt intake in many countries, including Australia,have been lacking [6, 7]. In the state of Victoria, salt in-take was estimated at 8.9 g/day in adults [8] and 6.7 g/day in children [9], with men and boys consuminghigher amounts than women and girls [8, 9] and bothadults and children exceeding the recommended max-imum salt intake [2]. To coordinate actions to reducesalt intake, in 2014 the Victorian Salt Reduction Partner-ship (referred to as the Partnership) was established bythe Victorian Health Promotion Foundation (VicHealth)

[10]. The Partnership, consisting of key organisations in-cluding VicHealth, The George Institute for GlobalHealth (TGI), Heart Foundation, Deakin University In-stitute of Physical Activity and Nutrition (IPAN), andthe Victorian Department of Health and Human Ser-vices (observer), developed a multi-component interven-tion strategy, informed by global evidence, to reduce theaverage salt intake of Victorians by 1 g per day by 2020[10, 11]. The Partnership project, which is described infurther detail elsewhere [12], comprised six main actionareas, including four intervention arms (consumerawareness campaign, generate public debate, food indus-try engagement, and advocacy and policy strengthening),building a strong partnership and a research and evalu-ation component [12, 13]. The Partnership comprisedthe Strategic Partnership group, an implementation teamand a research team (Fig. 1). The Strategic Partnershipwas tasked with providing strategic oversight to guideimplementation of collaborative action and met quar-terly throughout the intervention duration. The HeartFoundation was contracted by VicHealth to lead the de-livery of the four intervention arms, with support fromVicHealth and TGI through fortnightly meetings. Part-nership organisations, led by TGI, secured NHMRCfunding to conduct research and evaluate the interven-tion and met quarterly. Communication between Part-nership groups was primarily through VicHealth.

Fig. 1 Structure, roles and responsibilities of the Partnership and food industry stakeholders. Broken line indicates engagement with Partnershipthrough the intervention

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 2 of 21

To monitor and evaluate the effectiveness of the inter-vention, a comprehensive process and outcome evalu-ation was designed, as described in the protocol [12].The aim of the process evaluation was to examine thereach, dose, fidelity, context, adoption and effectivenessof the intervention. This was done through the collec-tion of routine administrative and cost data, impact as-sessments of consumer campaigns, product categoryreports, industry engagement, and advocacy activities,and analysis of stakeholder perspectives [12]. Semi-structured stakeholder interviews were conducted withkey stakeholders at two timepoints: (1) in the earlystages of intervention implementation (March to May2017 [14]) and (2) towards the end of the intervention(May to December 2019). This paper reports on the in-terviews undertaken in 2019. The primary objective wasto understand stakeholder perceptions of the effective-ness of the Partnership in achieving its intended roleand delivering a salt reduction intervention. Key lessonsand learnings generated from this research will be usedwith a view to contributing to greater understanding ofhow strategic partnerships can facilitate implementationof public health and nutrition interventions through col-laborative action and capacity building.

MethodsSample and recruitmentPotential participants were stakeholders involved in de-signing and/or delivering the Partnership intervention orengaged through the Partnership intervention, includingthe food industry, identified by VicHealth and the HeartFoundation. A letter of invitation to participate, along withthe participant information sheet and consent form, wasemailed to the VicHealth’s Partnership member contactlist in April 2019, and Heart Foundation food industrycontact list in August 2019. Participants agreed to partici-pate by return email, including returning a signed consentform and agreeing to an interview time. Interviews wereconducted from May to December 2019, either in personor online using Skype for Business or Zoom.

InterviewsSemi-structured interviews were conducted by two re-searchers, E.R., a PhD candidate and dietitian at TGIwho was not directly involved in the Partnership, andW-K.C., a research assistant at TGI who was not previ-ously involved in the project. The survey instrument waspreviously developed for interim stakeholder interviewsin 2017 to understand perceived barriers and enablers tointervention implementation [14] and was built on in-struments used in similar studies [15, 16]. Questionscentred on understanding the Partnership (effectiveness,structure and function), the process of executing theintervention, and internal and external factors affecting

intervention design, delivery and outcomes. Questionswere asked about each participant’s roles and their per-ceptions on delivery and fulfilment of their roles. Thesemi-structured approach allowed interviewers to adaptthe questions based on each participant’s involvement inthe Partnership, which enabled deeper knowledge in spe-cific areas. Probing questions were asked to gain moreinformation where necessary.Interviews were audio recorded using a phone and lap-

top. Permission to record was sought from participantsin the consent form and verbally at the beginning of theinterview. The interviews were manually transcribed byMurray Transcription.

Data analysisParticipants were characterised according to their in-volvement in the Partnership: Members of the StrategicPartnership (SP), implementation team (I) and/or re-search team (R); or food industry stakeholder engagedthrough the Partnership intervention (Fig. 1).Transcripts were de-identified and imported into NVivo

for data management. Transcripts were thematically ana-lysed by one researcher (E.R.), using a combination of de-ductive and inductive methods, with input from theresearch team (W-K.C., B.M., K.T. and J.W.). The Consol-idated Framework for Implementation Research (CFIR),was used to guide the qualitative analysis of the interviewsto better understand factors affecting program implemen-tation [17] within the context of the main process evalu-ation framework [12]. The CFIR is a comprehensiveframework designed for evaluations of interventions tounderstand the effectiveness of intervention implementa-tion and specifically “what factors influenced implementa-tion and how implementation influenced performance ofthe intervention” [17]. The CFIR consists of five domains:Intervention characteristics (e.g. design and developmentof the intervention), outer setting (contextual factors af-fecting intervention design, implementation and outcomese.g. political and social factors), inner setting (e.g. net-works and communication within the Partnership),characteristics of individuals (e.g. knowledge and beliefsabout the intervention) and the process of implementation(e.g. execution of the intervention). To enable a more in-depth analysis of the CFIR process domain, elements ofthe broader process evaluation framework were incorpo-rated [11]. Constructs included in this analysis were:Stakeholder perceptions on achieving intervention aims(e.g. reach and dose), and fidelity and quality of the inter-vention. The CFIR domains and adapted constructs usedin this analysis are displayed in Table 1. To ensure allmeaningful data were captured, an inductive approach,line-by-line transcript analysis, was used to identifythemes not already captured by the framework [18].

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 3 of 21

Ethics and consentThis study was approved by the University of SydneyHuman Ethics Research Committee (2016/770). Writteninformed consent was obtained from all participantsprior to the interview.

ResultsSampleTwenty-four stakeholders from 11 partner organisationsand 19 food industry stakeholders were invited to beinterviewed. Sixteen Partnership stakeholders agreed tobe interviewed, two declined as they felt they were notan active member of the partnership, one declined asthey felt a colleague (already invited) was best placed toparticipate, two were no longer with their organisation,and three did not respond. An additional two stake-holders agreed to participate however did not completethe interview process. Seven food industry stakeholdersagreed to be interviewed, two declined as they no longerworked for the company engaged by the Partnership,and ten did not respond.

In total, 14 Partnership stakeholders and seven foodindustry stakeholders were interviewed. Of the Partner-ship stakeholders interviewed, 12 were part of the Stra-tegic Partnership, five were part of the implementationteam and eight were part of the research team. Sevenparticipants were members of more than one group andtwo were members of all three groups (Fig. 1). Of theseven food industry stakeholders, one was employed bya major retailer and 6 were from large food manufactur-ing companies [19]. The average duration for Partner-ship interviews was 39 min and for industry interviewswas 30min. In general, food industry stakeholders wereonly able to speak about the industry engagement armand were not aware of the broader Partnership strategy.In the subsequent sections, we describe the key themes

from the stakeholder interviews using the CFIR adaptedconstructs with supportive quotes. Themes are organisedas follows: (1) the Partnership (inner setting, interven-tion characteristics and individual characteristics), (2)execution of the four intervention arms includingachieving intervention aims (e.g. reach and dose), en-ablers and barriers to execution, and fidelity and qualityof the intervention (process), (3) contextual factors af-fecting intervention design, delivery and outcomes (outersetting). An overview of the Partnership program, andstakeholders’ perceptions on the effectiveness of theintervention, is illustrated in the revised project logicmodel (Fig. 2).

The PartnershipPerceived effectiveness, structure and function of theStrategic PartnershipThe establishment of the Strategic Partnership was de-scribed as one of the biggest successes of the project.Stakeholders believed that the Strategic Partnership waseffective in achieving its goals of capacity buildingthrough the transfer of knowledge, skills and expertisebetween members, and collaborative action through thedevelopment and execution of a shared action plan. TheStrategic Partnership was viewed by most as essential forintervention planning and decision-making processesthat underpinned the execution of the intervention andperceived to be a “background enabler” for interventiondelivery (Table 2).Interviewees felt that having the “right” organisa-

tions and individuals involved, with a diverse range ofskills and expertise, and creating a positive learningclimate, where they were able to share ideas and pro-vide input into the strategy, were important compo-nents for designing and delivering a feasible,evidence-based salt reduction intervention in Victoria.Strong engagement from organisational leaders andsalt champions were enablers for establishing and sus-taining momentum and enthusiasm for the project.

Table 1 Consolidated Framework for Implementation Research(CFIR) domains and adapted constructs

Domains Adapted constructsa

Interventioncharacteristics

Intervention design and development

Relative advantage

Adaptability

Outer setting Cosmopolitanism

External policies and incentives: National level,state level

Other outer setting factors: Political context,socio-cultural factors, environmental factors,technological factors.

Inner setting Structural characteristics: Partnership structure,organisational roles and responsibilities,changes in personnel, organisational changes

Networks and communications: Collaboration,communication

Implementation: Compatibility, learning climate

Readiness for implementation: Available resources

Characteristics ofindividuals

Knowledge and beliefs about the intervention

Self-efficacy

Individual identification with Partnership

Process Planning

Engaging the right stakeholders: Opinion leaders,formally appointed internal implementationleaders, champions, external change agents

Executing the intervention: Achieving goals,enablers and barriers to delivery, fidelity,utility/quality

Reflecting and evaluatingaBold indicates additional construct

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 4 of 21

The commitment of members of the Strategic Part-nership, demonstrated through their consistent attend-ance at quarterly strategic meetings, was viewed as anenabler to achieving strong communication, joint de-cision making and collaborative action (Table 2).Most members spoke positively about the experience

of being involved in the Strategic Partnership and thedevelopment of the shared action plan. Though onestakeholder shared that there was some healthy disputebetween members. A few stakeholders questionedwhether the agreed strategy contained the right inter-ventions. They suggested that the intervention armsaiming to change the food environment, such as Indus-try Engagement, should have received greater focus andresources (Table 2).Differences in perspectives on the Partnership struc-

ture and function illustrated important communication

and knowledge gaps. Some of the implementation teamviewed the role of Strategic Partnership as less import-ant, likely because day-to-day functioning and decision-making was carried out at the level of the implementa-tion team. A few interviewees proposed this communica-tion gap was the result of a high turnover of staff withinthe implementation team and suggested more couldhave been done to transfer knowledge and capacity tonew implementation team members through better on-boarding processes (Table 2).These gaps resulted in a misunderstanding of the Stra-

tegic Partnership’s roles and responsibilities by some,which was illustrated in how interviewees describedStrategic Partnership members contributions to the co-ordinated efforts for delivering the intervention. Inter-viewees perspectives on the roles and responsibilities ofkey partner organisations are displayed in Table 3.

Fig. 2 Revised logic model of the Partnership program. Adapted from Trieu et al. [12]. Green box – stakeholders thought this was achieved.Amber box – stakeholders thought this was partially achieved. Red box – stakeholders thought this was not achieved

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 5 of 21

Table

2Stakeh

olde

rqu

otes

illustratingkeythem

esforeach

processelem

ent,by

actio

narea

Actionarea

,aim

andap

proach

Proc

esselem

ent

Them

eQuo

tes

Establishing

aStrategicPartne

rship

Aim

:tocreate

aStrategicPartne

rshipandde

velopa

collabo

rativeapproach

tosaltredu

ctionwith

stakeh

olde

rs

App

roach:VicH

ealth

invitedstakeh

olde

rorganisatio

nsto

join

thePartne

rshipandde

velopajointactio

nplan

forsalt

redu

ctionin

Victoria.

Perspe

ctives

onachievingaims

Learning

how

towork

toge

ther

It’satrue

partnershipwhereeveryorganisationbrings

streng

thsto

thetable...the

realsuccessof

thisprojectisthat,learningho

wto

collaborate

andworkwith

others

andrecogn

iseeach

ofthestreng

thsthat

they

bringisreallyimportan

t.(M

ember9:

SP,R)

Achieving

collabo

rativeactio

nIthink

thepartnershipha

dachieved

wha

titsetoutto

achievein

term

sof

collaborativeactionan

dsettingof

actionplan

that

isaboutus

working

togetherfor

saltreductionspecificto

thestate.(M

ember11:SP,I,R)

Achieving

capacity

building

Ithink

thereha

sbeen

good

ongoingdiscussio

n,collaboration,Ithink

there’sbeen

abunchof

capacitybuilding.(M

ember4:SP,R)

Enablersto

interven

tionde

livery

The“right”Partne

rsIthink

youha

dtheright

organisations,stron

gleadershipwithin

thoseorganisations

andgood

expertise.(Mem

ber2:SP,I,R)

Com

mitm

entfro

mPartne

rsIthink

that’sbeen

criticalinha

ving

allo

ftho

sepeoplewith

acommon

interest

around

thetablebecauseit’sbeen

such

along

time,theworking

groupperiod,an

dyouneed

arang

eof

playerscommitted

tothat

tokeep

mom

entum

goingthrough.

(Mem

ber1:SP,R)

Regu

larstrategicmeetin

gsTheabilityto

have

ameetingeverythreemon

ths,samepeoplecomingtogether,

who

heardtheconversationlastmeeting,at

that

partnershiplevel–

itusually

makes

itmucheasieran

dit’skind

ofan

enabler.(M

ember11:SP,I,R)

Theregularmeetings

werereallygood

opportun

ities,theywerewell-run

andfocused

andso

Ithink

they

werefantastic

opportun

ities

tobuild

mom

entum

inthesaltspace

andthehypertensio

nspace,wereallyha

dn’tha

dsomething

likethat

forareally

long

time,so

that

was

terrific.(M

ember8:SP)

Diverse

skillsandexpe

rtise

(capacity

building)

Ithink

thosecoregroups

brough

tareallygood

mixof

skillsan

dexpertise,w

hich

Ithinkwealllearntfro

m.Inmyearly

days

onthepartnershipwhenwe’reestablish

ing

theagenda,itwas

reallyusefulan

dgreatbecausewealllearned

from

onean

other

andyouknow

weha

dsomeprettyfierykind

ofdebatesan

ddiscussio

ns,w

hich

were

great.(M

ember7:SP)

Develop

ingashared

actio

nplan

Lookingat

theevidence,eng

agingwith

thekeystakeholdersaround

it,appraisin

goptions

foraction,feasibility,politicala

cceptabilityawho

leho

stof

different

domains

tothen

draw

-upashared

plan

,onwha

ttheconsensusforactionon

that

wouldbe

that

everyone

couldco-com

mitto.(Mem

ber13:SP)

Barriersto

interven

tionde

livery

Less

activeinvolvem

entfro

msomepartne

rsIthink

we’ve

hadsomewho

haven’tactivelycontributed

inthesameway

asothers...

they

shareourmessagesan

dourcampaign,butitwouldn’talwaysfeellikethey

had

astake.(M

ember11:SP,I,R)

Toomanype

ople/

organisatio

nsYoudo

have

togetalotof

peopleinvolved

andalotof

peopleacrossthings

orsig

nedoffb

ypeople.(Mem

ber5:SP,I)

Slow

start-up

Itmight

have

been

agood

year

beforeitwas

quite

clearw

hatthey

wereplanning

todo

tome…

tryingto

desig

nimplem

entationstrategies…

lookingat

gettingother

stakeholdersinvolved

anditisquite

aslowprocess,butittook

awhileform

eto

clearly

understand

whatthe

implem

entationstrategies

weregoingto

be.(Mem

ber3:R)

Transfer

ofknow

ledg

eand

skillsto

new

mem

bers

Wecouldha

vedone

moreinterm

sof

reallymakingsurethenewpeoplewho

came

inwerefully

awareof

wha

ttheprecise

goalsof

thepartnership,thewaysof

working

andthat

wedideverything

wecouldto

ensurethat

thecapacitywas

tran

sferredfor

therelevant

people…Ithink

sometimes

youjustun

derestimatethebenefitsor

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 6 of 21

Table

2Stakeh

olde

rqu

otes

illustratingkeythem

esforeach

processelem

ent,by

actio

narea

(Con

tinued)

Actionarea

,aim

andap

proach

Proc

esselem

ent

Them

eQuo

tes

stopping

,takingstock,makingsurethat

absolutelyeverybodyin

thepartnershipis

onthesamepage

andha

sallthe

relevant

background

inform

ationan

dknow

ledge

andskillsin

orderto

ensurethat

theprojectcontinueson

thesametrack,an

dIthink

potentially

wecouldha

vespentsomemoretim

edoingthat.(Mem

ber2:SP,I,R)

Barriersto

achieving

outcom

esThe“right”interven

tion

Ithink

theHeartFoun

dationto

theextent

that

itdelivered

anintervention,did

deliveran

interventionprettyeffectively,butwhethertheinterventionwas

the

interventionthat

needed

deliveringIthink

isakeyquestion.(M

ember4:SP,R)

Everyone’spassiona

tewho

was

around

thetableon

thisissue

andof

course

you’re

goingto

have

someheat,and

that’spartof

it,itshouldn’talljustbe

smooth

sailing

,youneed

peopleto

question,challeng

ewhetheryouneed

theconsum

erpieceor

doyoujustfocuson

thefood

reform

ulationor

doyoudo

thedebate

andthepolicies?

(Mem

ber7:SP)

Perspe

ctives

onthe

fidelity

ofthe

interven

tion

Chang

esto

Partne

rship

mem

bers

Anu

mberof

otherplayersha

vecomeon

asit’sdeem

edrelevant

that

they

can

contribute,an

dobviously

they

have

aninterestin

thisarea.(Mem

ber1:SP,R)

Perspe

ctives

onutility/qu

ality

ofthe

interven

tion

EffectivePartne

rship

It’sbeen

reallyeffectiveto

have

thesedifferent

organisations

with

different

skillsets

working

socloselytogether.(Mem

ber21:SP,I)

Enablerto

interven

tion

delivery

ThePartnership’sabackground

enabler.(M

ember12:SP)

RaisingCon

sumer

Awaren

ess

Aim

:toincrease

publicaw

aren

essof

saltintakes,saltand

health,d

ietary

sourcesof

saltandstep

sto

redu

cesaltintake.

App

roach:athree-ph

asedigitalcam

paignthat

targeted

pri-

maryfood

providerswith

infamilies

with

prim

aryscho

ol-

aged

children.

Perspe

ctives

onachievingaims

Positiveachievem

ents,g

iven

limitatio

nsWith

thecampaign,they’vedone

well,giventhebudget

andallthe

otherfactors,

that

was

alwaysgoingto

makethat

atoughtask.(Mem

ber10:SP)

Goo

dreachandrecallwith

inthetarget

audien

ceNearly

50%

ofthetarget

marketswerereachedwith

thiscampaignacrossthethree

yearsthat

wedidan

intervention,which

ishu

ge,and

then

ofthat,82%

ofthat

are

recalling

oneof

ourtopkeymessagesaboutthat

there’sprocessedfood

ishigh

salt,

that

they

need

tobe

readinglabelsan

dlookingforlowersaltoptions,and

eating

fresh

isbest.So,there’ssomereallygood

positiveoutcom

esan

dthey

were

achievem

entsagainstobjectives.(Mem

ber21:SP,I)

Limitedim

pact

onge

neral

popu

latio

nThemetricswe’ve

goton

awareness-raising

wereprom

ising

andthat’sgreatbut

Deakin’sresearch

wouldsaywe’reno

tchan

ging

know

ledge,attitudes

andbeha

viours

onageneralpublic

levelbut

potentially

with

that

target

marketof

campaignwe

have

raisedaw

areness.(M

ember11:SP,I,R)

Con

certed

campaign

Wewereableto

runaconcertedcampaignforfour

yearsaround

gettingVictorians

toreduce

theirsaltintake

-Ithink

that

was

ahu

geachievem

ent.(M

ember9:SP,R)

Enablersto

interven

tionde

livery

Utilisingearntmed

iato

increase

reach

Whilst

allour

paidadvertising

focusedon

Victoria,allof

ourorganicsocialan

dearnt

mediawas

alln

ationa

l,so

wereachedfarbeyond

that.(Mem

ber21:SP,I)

Barriersto

interven

tionde

livery

Limitedmed

iums

Look

Ithink

[thereach]

islim

itedto

somedegree

becauseit’son

lysocialmedia…

there’sno

broaderbrush...w

eha

ven’tgotadiversity

ofmediums.(M

ember7:SP)

Weakmessaging

Our

callto

actionin

ourconsum

eraw

arenessha

sperhapsno

tbeen

asstrong

asit

could’ve

been.(Mem

ber7:SP)

Barriersto

achieving

outcom

esInability

tocut-throug

hothe

rnu

trition

message

sIthink

thetarget

audience

that

weweretargetingwas

problematicinthat

they’re

beingbombarded

byallsortsof

messagesaround

wha

tto

feed

theirkids

andho

wto

feed

theirkids

andho

wto

getthem

exercisin

gor

howto

maintaintheirhealth.

So,Ithinkthereweresomechalleng

esin

perhapsthetarget

audience

that

we

selected.(Mem

ber9:SP,R)

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 7 of 21

Table

2Stakeh

olde

rqu

otes

illustratingkeythem

esforeach

processelem

ent,by

actio

narea

(Con

tinued)

Actionarea

,aim

andap

proach

Proc

esselem

ent

Them

eQuo

tes

Target

audien

cedidno

tsee

therelevance

Ithink

we’vehadproblemswith

thetargetaudience

forthe

consum

ergroupandI

don’tknowwhetherthey

werethebestplaceto

start...You

canoftenfindsaltandthat

issue

with

theoldera

gegroupbecausethey

seeitas

relevant

tothem

andwe’veseen

that

throughtheconsum

eraw

arenessthat

thepeoplethat

werethebiggestu

ptakeare

thosethat

areolderb

ecause

they

seetherelevance.(Mem

ber7:SP)

Shortinterven

tiontim

eframe

Allcreditto

thepartnersan

dthePartnershipin

that

wewereableto

dothisforfour

yearsbutshiftingconsum

erbeha

viouran

dchan

ging

anentirefood

supplyisareal

long

-term

commitm

entan

dyou’reno

tgoingto

seesig

nifican

tshiftsin

thespaceof

four

years.(M

ember9:SP,R)

Perspe

ctives

onthe

fidelity

ofthe

interven

tion

Decisionto

tryandmovethe

target

audien

cealon

gthe

Stages

ofChang

e

Atfirst,w

esortof

wan

tedto

makethem

awarean

dsay,“look

thereisalotof

salt

inthesepackaged

products,sowewan

tyouto

beaw

areof

that”.An

dthen

they

werelike“okay,maybe

wearemakingthem

aware,weha

vedone

afewdifferent

campaignbursts,can

wegetthem

toconsiderthisabitmore,considerdoing

something

aboutthis”.Sothen

that

was

introduced,m

oreof

likepotentially

asw

apmessage,and

then

bythetim

eIcam

eto

itlastyear

whenIstarted,through

discussio

nswith

thegeneralw

orking

groupan

dthePartnershipgroup,itwas

sortof

like,“look,w

hydon’tweseeifwecanpush

peopletowards

that

beha

viourchan

geso

canweactuallygetthem

toact,canweactuallygetthem

todo

something

aboutreducing

theirsalt”.(Mem

ber5:SP,I)

Reverted

toaw

aren

essand

considerationdu

eto

cost-

efficiencies

Wedecidedto

cutthat

actionan

dacquisitionpartan

djustfocuson

awarenessan

dconsideration,becauseweknew

with

thebudget

weha

d,wecouldstillretain

and

getgood

numbersan

dhealthytraffic

andcost-efficiencies

ineducatingpeopleat

that

stage(M

ember5:SP,I)

Limitedmed

iumsafterfirst

campaignroun

dThefirstroun

dof

thecampaign,for“Unp

acktheSalt”,w

edidoutdooradvertising

,which

perfo

rmed

well.Wedidn

’tha

vequite

asmuchmon

eythesecond

timeweran

it,so

itwas

then

morejustdigitaland

online...Whenyouchan

gebecauseof

finan

ces,youmight

lose

alittle

bitof

your

impact.(Mem

ber11:SP,I,R)

Perspe

ctives

onutility

/qu

ality

oftheinterven

tion

Not

effectivein

increasing

publicde

mand

Wha

twewereho

ping

was

that

thecampaignoriginallywouldgenerate

someof

thecommun

ityaw

arenessan

dactionto

helpelevate[salt]as

aconversation,abit

likeitha

swith

family

violence

andisstartingto

happen

with

obesity.Tha

tha

sn’t

reallyha

ppened

andso

thepoliticians

have

been

ableto

kind

ofigno

reitforwan

tof

another.(M

ember10:SP)

Not

effectiveat

popu

latio

nsaltredu

ction

With

thesm

alln

umberof

peoplethat

that

impacted

then

that

was

agood

respon

se,but

itwas

asm

alln

umberof

peopleso

atapopulationlevelIwouldsayit

was

ineffective(M

ember10:SP)

Gen

eratingPu

blicDeb

ate

Aim

:toge

nerate

publicdiscou

rseandde

bate

arou

ndsalt

redu

ction

App

roach:med

iareleases

dissem

inatingprod

uctcatego

ryrepo

rtsthat

high

lighted

saltlevelsin

processedfood

sthroug

hout

theinterven

tionpe

riod.

Perspe

ctives

onachievingaims

Saltin

thepu

blicdo

main

Ithink

we’ve

hadsalttalked

aboutin

thepublicdomain,prettyconsistently,and

Ithinkthat’sareallycompetitiveenvironm

entfornu

trition

,soIthink

that’sapretty

greatachievem

ent.(M

ember11:SP,I,R)

Shapingpu

blicde

bate

Ithink

theresearch

hasbeen

particularlypowerfuland

productive.It’sbeen

amajor

partof

thepartnershipan

dIthink

thequality

oftheresearch

comingout,an

dthe

way

inthat’sbeingused

totryan

dshapepublicdebate,has

been

high

lyeffective.

(Mem

ber13:SP)

Enablersto

interven

tionde

livery

Nam

ingcompanies

Thoseresearch

reports,theway

that

they

weredesig

nedto

actuallyna

me

companies

andbran

ds,m

eant

wegotgreatmediapick-up,that

meant

wegotgreat

reach.(M

ember11:SP,I,R)

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 8 of 21

Table

2Stakeh

olde

rqu

otes

illustratingkeythem

esforeach

processelem

ent,by

actio

narea

(Con

tinued)

Actionarea

,aim

andap

proach

Proc

esselem

ent

Them

eQuo

tes

Barriersto

interven

tionde

livery

Alignm

entof

organisatio

nal

andPartne

rshipview

sTheHeartFoun

dationis,

asaremostorganisations

areno

w,shifting

away

from

the

nutrientsfocus…

toamorewho

le-fo

odapproach.So,wecertainlyyesfeltsometen-

sions

there…

whenwe’relookingat

food

categories(M

ember21:SP,I)

Someof

thefood

categoriesthat

VicHealth

andtheGeorgeInstitute

wouldha

veliked

togo

outwith

didn

’tsit

wellw

iththeHeartFoun

dation’sph

ilosoph

y,so

[the

Partnership]

couldn

’tpursue

thoseopportun

ities.(Mem

ber21:SP,I,R)

Barriersto

achieving

outcom

esCrowde

dnu

trition

space

Thenewspapersarefullof

debatesabout‘cleaneating,’and

caffeinewillgive

you

thisan

dyour

processedfatswillgive

youthat,w

eused

totalkaboutfatsan

dno

wwe’retalkingaboutsugars.Inthat

context,talkingaboutsaltitwas

alwaysgoingto

struggleto

getcutthrough.(M

ember13:SP)

Perspe

ctives

onthe

fidelity

ofthe

interven

tion

Chang

esin

thede

sign

ofthe

interven

tion

Ifweno

wwentinto

desig

ning

this,

we’dgo

inwith

theintentionof

implem

entation

lookinglikeitendedup...Ittook

awhileto

getto

that...thisidea

ifyoubenchm

ark

industryaround

anu

trient

oraround

thehealthinessof

theirproducts,you

can

leverage

them

forpublicaw

arenessan

dfood

industryengagementan

dpotentially

governmentpolicyengagementdepend

ingon

wha

tthepoliticalland

scapeis.

That’s

akeylesson

.(Mem

ber11:SP,I,R)

Perspe

ctives

onutility/qu

ality

ofthe

interven

tion

Engaging

med

ia,ind

ustryand

policymakers

Wha

t’sbeen

reallyeffectiveha

sbeen

theproductcategoryreports,which

theGeorge

Institute

hasdone.These

regularreports,which

basicallylook

atsaltlevelsof

certain

foods,weusethoseforstrategicadvocacy

togetmediaattention...itraises

awarenessof

thepublican

dourpolicymakers,butitalso

enablesus

toliaise

directly

with

thefood

companies

toraise

theiraw

areness...and

invite

them

tothetable,an

dto

developstrategies

totryan

dreduce

salt...that’sworkedrelativelywell.(M

ember

2:SP,I,R)

Scalability

Thisconceptof

doingthesesurveyson

food

grou

psan

dresurveyingeveryfew

yearsan

dthat

namingan

dsham

ingof

individu

alprod

ucts,brilliant,

absolutely

criticalan

dIreckon

that

was

something

that

wou

ldbe

scaled

upan

dareally

valuab

lepieceof

work.

Ithou

ghtthat

approa

chwas

very

impo

rtan

t.(M

ember8:

SP)

Inno

vativeapproaches

with

food

indu

stry

Aim

:Toredu

cesaltlevelsin

packaged

andprocessedfood

sby

supp

ortin

gfood

manufacturersto

reform

ulate.

App

roach:Engage

food

indu

stry

throug

heven

tsand

meetin

gs,d

evelop

resources,services

andgrants,and

prod

ucecase

stud

ies.

Perspe

ctives

onachievingaims

Establishing

relatio

nships

We’ve

been

successfulin

developing

reallystrong

relationships

with

someof

Australia’smajor

food

man

ufacturers(M

ember6:I)

Und

erstanding

reform

ulation

practices

Wesetoutto

engage

with

topman

ufacturersan

dun

derstand

wha

t’sha

ppening

within

thesebusin

essesan

dthen

toshow

case

theirpractice,where

itwas

available

andtryan

dhelpthosewho

aren’t,an

dIthink

weha

vedefinitelyachieved

allof

thosethings

(Mem

ber21:SP,I)

Develop

ingsupp

ortive

resourcesandservices

Andwealso

wereableto,asaresultof

working

with

[thefood

industry],developa

guideto

supportparticularlytheVictorian-basedobjectives

that

weoriginallyha

dat

theoutset

oftheproject.So

that

guidewas

something

wecoulddevelopforthese

smallerto

medium

man

ufacturersbasedin

Victoriaan

dthat’sbeen

shared

with

them

throughvarious

networks.Soyes,Ithink

we’ve

definitelyachieved

thoseout-

comes.(Mem

ber21:SP,I)

Chang

esto

thefood

supp

lyHow

effectivewillactivity

that

haskind

ofsped

upin

thelastyear

ortwobe

inactuallygettingsaltlevelsdownin

thetim

ethat

we’remeasuring?...Ifyoulooked

atitright

now,I’m

notsureifwecouldseethat

thefood

supplyha

salreadyshifted.

(Mem

ber1:SP,R)

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 9 of 21

Table

2Stakeh

olde

rqu

otes

illustratingkeythem

esforeach

processelem

ent,by

actio

narea

(Con

tinued)

Actionarea

,aim

andap

proach

Proc

esselem

ent

Them

eQuo

tes

Enablersto

interven

tionde

livery

Buildingpo

sitiverelatio

nships

Your

overalla

imwas

toreallysupportman

ufacturersan

dyouwan

tedto

provideus

with

thereform

ulationguide,youha

dan

eventthat

was

really,really

supportiveof

companies

who

weredoingthings

welland

that

reallyhelped

usto

getinvolved.

(Industry15)

Barriersto

interven

tionde

livery

Uncertainty

ofrolesand

authority

Thisisha

ppeningat

Federallevelan

dthereforewha

tisourrolean

dwha

tisour

authority

interm

sof

industryengagement.So,Ithinkitwas

that

lack

ofclarity

ofroles,an

dwheredo

wefit

andwha

t’sourauthority,Ithinkthat

was

probablyon

eof

thebarriers.(Mem

ber2:SP,I,R)

Slow

startto

implem

entatio

nEngaging

with

food

industryat

thestartmay

have

been

abitslo

wbutIpersona

llydon’tha

veexperienceengaging

with

food

industry.O

therpeopleat

thetableha

vesome,butIthink

it’stakenawhileto

probablycomeup

with

theworking

plan

ofwha

tisthebestapproach

toengage

with

that

target

group.(M

ember1:SP,R)

Perspe

ctives

onthe

fidelity

ofthe

interven

tion

Chang

esto

theplansdu

eto

new

know

ledg

eand

inno

vatio

n

OverallthePartnershipha

sbeen

soopen

totrialand

erroran

dthelevelo

finno

vationin

term

sof

let’s

trythisan

dlet’s

trythat.The

resourcesthat

have

been

developedby

thepartnershipforindustry,w

eneverenvisagedthat

atthebeginn

ing.

Who

wouldha

veenvisagedtwoyearsagothat

wewouldha

vefinish

edthisproject

with

aho

wto

guide,benchm

arking

reportsan

dgran

ts?(M

ember6:I)

Chang

esto

thetarget

audien

ceto

effect

more

change

Wechan

gedha

lf-way

throughinregardsto

ourthinking

around

who

weshouldtar-

get....Ifweha

dbeen

thinking

atthestartaround

howcanweeffectthemost

chan

gean

dtarget

thebigon

eswith

thesefood

categoryreportsan

dthen

howdo

weeffectthemostchan

gein

Victoriaby

providingtoolsan

dresourcesan

dcontacts

forsm

alltomedium,tha

twouldha

vebeen

anice

roun

dbran

dfro

mthestart.

(Mem

ber21:SP,I)

Perspe

ctives

onutility/qu

ality

ofthe

interven

tion

‘ReformulationReadiness’

Guide

Thereform

ulationreadinessguide...Ithink

that

isagreatresource

that

was

putout...

Weha

dalotof

good

feedback

from

different

areasof

oursupplychainan

dfood

techno

logistsaround

that

guide,so

justreallyhelpful.(Industry17)

Indu

stry-cen

tred

even

tsIt’sprobablythefirsttim

eweha

veseen

apublichealth

organisationrunan

industry

eventan

dbe

supportive...youun

derstand

theworkwearedoing,an

dyou’rewilling

tosupportitan

dhigh

light

wherecompanies

aredoingwell(Industry15)

Casestud

ies

Smallerbusin

esses...theyneed

toseewha

twe’redoingan

dthey

need

tolearnfro

mus

andifthey

seecase

studiesfro

mus,tha

tcanreallyhelp

them

shapetheir

nutrition

strategy,soIthink

it’sapositiveforthoseotherindustryplayersan

dalso

consum

ersto

seewha

twe’redoingan

dobviously

governmentto

seewha

twe’re

doing(Industry17)

Benchm

arking

service

TheGeorgeInstitute

was

offeringthosereportswhich

couldtake

onecompanies’

productsan

dcompareitto

theentirecategory...Ithink

that’sreallyvaluableas

well

becauseyouknow

,allcompanies

aregoingto

beinterested

toknow

wha

ttheir

competitorsaredoing.(Industry16)

Reform

ulationgrants

Thesaltreductiongran

ts,Ithinkthat’sshow

ingreallystrong

leadershipan

dshow

ing

food

man

ufacturersthat

thereissupportan

dthereareresourcesavailableto

dothis.

(Mem

ber6:I)

One

-to-on

emeetin

gsWe’ve

been

able

toun

dertakeover

20on

eon

oneenga

gementmeetin

gswhich

have

revealed

alotof

insigh

tan

dinform

ationarou

ndthedriversfor

reform

ulationan

dthecurrentob

staclesan

dissues

that

food

man

ufacturers

are

still

facing

.(M

ember6:

I)

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 10 of 21

Table

2Stakeh

olde

rqu

otes

illustratingkeythem

esforeach

processelem

ent,by

actio

narea

(Con

tinued)

Actionarea

,aim

andap

proach

Proc

esselem

ent

Them

eQuo

tes

Advocacyandpo

licystreng

then

ing

Aim

:to“stren

gthe

nhe

althypo

licies”to

improvethefood

environm

ent,includ

ingfood

reform

ulationprog

ramsand

institu

tionaln

utritionpo

licies.

App

roach:Engaging

governmen

tstakeh

olde

rs,creatinga

policypo

sitio

nstatem

entandinfluen

cing

governmen

tinitiatives.

Perspe

ctives

onachievingaims

Did

notachievepo

licychange

Ireckonifwe’djustgone,“Youknow

wha

t,to

makean

impactin

Victoriawha

twe

need

todo

isweneed

tochan

gethings

nationa

lly,”wejustspentallofo

urtim

ean

deffortjustfocusin

gon

bringing

theVictorianStategovernment,regulators,

poweran

dinfluence

tobear

onCa

nberra

andindeed

ontheotherstates,toactually

chan

gethings

forthewho

lecoun

try,maybe

that

wouldha

vebeen

moresuccessful.

(Mem

ber4:SP,R)

Whilst

wemight

notha

veseen

policychan

ge,w

e’ve

definitelycontinuedthe

conversationan

dputsupportbehind

it.(M

ember21:SP,I)

Supp

ortedalreadyplanne

dactivities

Ifeelliketheadvocacy

activities

probablyha

vejustaddedastrong

supporterbase

around

things

that

werein-train.(Mem

ber11:SP,I,R)

Enablersto

interven

tionde

livery

Develop

ingablueprint

Wedevelopedthecallto

actiondocument.Ithink

was

agreatoutput

anda

deliverable...it’sprobablythefirst-timeorganisations

have

gottogetherto

actually

getsomeform

ofconsensus,ablueprinton

wha

tweneed

goingforward.(M

ember

7:SP)

Barriersto

interven

tionde

livery

Lack

ofallocatio

nof

rolesand

respon

sibilities

Wedevelopedaclearstatem

ent,which

outlineswha

tthekeyasks

ofthestrategic

partnershipwere,butthen

wedidn

’tidentifypeerrolesor

allocate

specific

respon

sibilitiesto

thedifferent

partnershipmem

bersin

term

sof

who

was

goingto

take

that

forward.(M

ember2:SP,I,R)

Uncertainty

arou

ndow

nership

andleadership

Ithink

oneof

thechalleng

esof

thePartnership,whenyou’ve

gotmultiple

organisations,a

lotof

streng

ths,butthen

you’ve

gotacertaindegree

ofwho

’sow

ning

it,leading,drivingit,an

dwho

’sgoingto

makethosecalls

andwho

’sgoing

togetthecreditforit.Which

isun

fortun

atelythepoliticsworking

anyw

herean

dI

thinkthat’sprobablyheldsomeof

theprogressback

inthat

policyadvocacy

space

becausemixed

inwith

alltha

tpoliticalcontext.(M

ember10:SP)

Barriersto

achieving

outcom

esPo

liticalclim

ate

There’sbeen

somegood

workbutas

awho

leitprobablyha

sn’tquite

hadthe

impactat

thestatean

dfederallevelas

wewouldha

vewan

ted.That’sno

tnecessarily

becauseof

thefaultof

anyof

thepartners,it’s

partlybecauseof

the

politicalconversations

andagendasoutwheresaltisan

dyoucan’tmakean

issue

popularwith

politicians

ifthey

don’twan

titto

bean

dthere’sno

tapublicpush.

(Mem

ber10:SP)

Perspe

ctives

onthe

fidelity

ofthe

interven

tion

Shift

from

targetingstate

governmen

tto

fede

ral

governmen

t

Upun

tilprobablyJune

2018,w

ewereadvocatingat

astatelevel,an

dthen

we

shifted,w

ethough

titwas

moreof

ana

tiona

l,afederalissue...so

wewrote

anew

plan

(Mem

ber21:SP,I)

Attheverybeginn

ingof

theproject,itwas

veryVictoriafocused…

butalon

gthe

way

thedecisio

nwas

made…

it’sana

tiona

lthing

,ifthey’re

goingto

makethat

policy,it’sgoingto

beFederalpolicy.(M

ember5:SP,I,R)

Perspe

ctives

onutility

/qu

ality

oftheinterven

tion

Jointadvocacy

Therearelotsof

benefitsto

apartnershipworking

togetheron

strategicadvocacy

andIthink

we’ve

utilisedthat

quite

wellonsomeoccasio

ns,but

perhapsno

tas

well

aswecouldthrougho

utthewho

leproject,an

dIthink

thisisprobablyachalleng

eforthesesortsof

partnerships

ingeneral.(M

ember2:SP,I,R)

Advocacyasks

documen

tSomepeoplemight

saywelltha

t’sthestatem

entof

theobviousbutto

have

itas

adocumentthere,to

hand

-out

togovernmentsan

dto

useas

andadvocacy

toolwas

amileston

e.(M

ember8:SP)

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 11 of 21

Execution of the four intervention armsThe intervention action areas, process domains, keythemes and stakeholder quotes relating to the executionof the intervention are displayed in Table 2.

Consumer awareness campaignStakeholder perceptions on the effectiveness of the Con-sumer Awareness campaign were mixed. Stakeholders in-volved in, or overseeing, intervention delivery sharedthat a digital campaign was strategically chosen for itscost effectiveness and efficiency in reaching the targetaudience. Eight interviewees expressed that the reachwas “adequate” given the constraints of the chosen ap-proach, including the budget, timeframe and target audi-ence. Some of the intervention team shared specificachievements of the campaign, such as high recall of thekey messages within the target population (Table 2).However, overall, many felt the campaign was limited

and did not have enough reach. Two participants statedthat the campaign was ineffective in making a meaning-ful contribution to the overall Partnership goal of a one-gram reduction in population salt intake. Amongst thosewho expressed scepticism towards the campaign’s effect-iveness, challenges in delivering this intervention armand engaging the targeted population were discussed.Five reasons were suggested by interviewees: (1) The tar-get audience did not see the relevance for them becauseof the perception that high blood pressure was an olderperson’s problem; (2) the target audience was “bom-barded” with digital health messaging from othersources; (3) absence of hard-hitting campaign messagingthat did not “cut-through” other digital health messagesthat flooded media and social media spaces; and (4) lim-ited campaign mediums (digital only) as a result of tight

budgets; and (5) the intervention timeframe was tooshort (Table 2).

Generate public debateThe primary mechanism used to generate public debatewas perceived by interviewees to be an effective centrallever for consumer awareness raising, engaging the foodindustry and pursuing advocacy and policy asks throughmedia advocacy strategies. One stakeholder shared howthe strategy was developed overtime and many inter-viewees didn’t differentiate it from the ConsumerAwareness arm. Ten Partnership members discussed theutility of the product category reports in engaging mediaand industry. The reports were perceived to be a “starperformer”, achieving more media and industry engage-ment than anticipated, and at low cost (Table 2).However, the majority of interviewees (10 Partnership

stakeholders and three food industry stakeholders) alsoacknowledged a competitive media environment, whichwas “full of debates about ‘clean eating’” and “sugar”, assomething that potentially hindered impact. Implemen-tation team members identified that some individualsand organisations had changed focus from salt as a sin-gle nutrient towards general healthy eating principles, asa challenge that had created a misalignment between in-dividual, organisational and Partnership objectives andvalues resulting in friction between some members.These “tensions” were resolved; however, delivery of thisintervention arm was slowed and perceived as “challen-ging” at times (Table 2).

Food industry engagementThe Industry Engagement strategy was viewed as effect-ive in terms of innovative approaches to industry en-gagement and creating positive relationships between

Table 3 Organisational roles and responsibilities, as described by interviewees

Organisation Roles and responsibilities

VicHealth VicHealth has clearly set direction and part of the reason we’ve had that role is because we have funded theimplementation. (Member 11: SP, I, R)

Heart Foundation Heart Foundation was awarded a contract by VicHealth to deliver three areas of the intervention, being consumerawareness, advocacy and industry engagement (Member 21: SP, I)

The George Institute The George Institute was engaged for strategic advice and also as a research partner and subsequently built anNHMRC partnership grant around the work which the partners were engaged on (Member 13: SP)As we move forward, we have been supporting the Heart Foundation in terms of the intervention’s strategy.(Member 2: SP, I, R)

IPAN at Deakin University Much of our work has been collecting that baseline and end data for surveys in regards to knowledge, attitudesand behaviours have shifted on salt consumption throughout after this intervention, as well as actual salt intakein children and adults. (Member 1: SP, R)

Department of Health and HumanServices Victoria

Providing strategic advice and promoting the project broadly… contributing funding… government rep. (Member12: SP)

Other organisations on the StrategicPartnership

Attend four meetings a year, we’re just looking for their input, looking for their contribution, looking for where theymight be able to amplify communication messages (Member 11: SP, I, R)All of the aspects of the partnership plan were developed through consultation with myself and the other strategicleaders. (Member 13: SP)

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 12 of 21

public health organisations and the food industry. How-ever, both strategic partnership members and programimplementers identified challenges, including perceivedchanges to the strategy during the intervention (Table 2).Stakeholders cited a slow start to implementation of thisintervention arm. A few suggested this was due to aninitial lack of understanding of the food industry, how toengage them and which organisation would be respon-sible. Some interviewees said that the focus initially wason supporting small-to-medium manufacturers inVictoria to reformulate processed foods, but this waslater expanded to include national and internationalmanufacturers, in order to have an impact on the Aus-tralian food supply (Table 2).Stakeholders spoke positively about the learning ex-

perience and the process of adapting the strategy to ad-dress industry needs identified throughout theintervention. Both Partnership members and food indus-try stakeholders spoke of the strong positive relation-ships that were created. One-to-one meetings were seenas important for building rapport and a two-way ex-change of knowledge between the Partnership and foodindustry. Interviewees discussed how through thesemeetings, new knowledge and understanding of industrylevers and capabilities were gained, which enabled thePartnership to adapt the industry engagement strategyto utilise levers and create services and resources to fillidentified gaps in capabilities. This included: (1) the de-velopment of a guide for reformulation to support com-panies to undertake reformulation, (2) a benchmarkingservice to assess the nutritional profile of products andcompare to competitors and to sodium targets, and (3)grants for small-to-medium manufacturers to financiallysupport reformulation. In addition to increasing cap-acity, interviewees shared that the Partnership was ableto demonstrate supportiveness of companies’ positiveprogress towards sodium reduction through public casestudies and showcases at events. This was perceived tofurther facilitate relationship building and help overcomethe “us and them mentality” between public health orga-nisations and the food industry (Table 2).

Advocacy and policy strengtheningStakeholder perspectives on the effectiveness of the Advo-cacy and Policy Strengthening arm were mixed, likely dueto differences in understanding of the intended objectivesand strategy. Two program implementers shared how thepolicy and advocacy activities were initially carried out ata state-level, however the Partnership identified that keylevers for certain food policies were at different levels ofgovernment and that there was a mis-match between whatwas planned and what was achievable at a state-level. Spe-cifically, stakeholders spoke about food reformulation pro-grams being a federal matter and many suggested this

advocacy could have been more successful if it had a na-tional focus from the outset, though some recognised thechallenges associated with the funding coming from thestate for a Victorian-specific intervention (Table 2). Con-trastingly, one strategic member spoke of the plannedstrategy as always intending to target both the federal andstate governments and two strategic members shared thatvery little progress was made towards integrating salt re-duction into state-based policies, including institutionalnutrition policies.Participants talked about the development of an ad-

vocacy asks document, that was perceived to be vitalin bringing the organisations together and forging aconsensus on the way forward for the Partnership.However, the utility of the document, and indeed theoverall strategy, was a contested topic, with someviewing it as a “milestone” and others a “statement ofthe obvious”. Many shared how the Partnership had“affected things that have been going on”, such asproviding feedback on the draft sodium reduction tar-gets proposed by the Australian federal government’sHealthy Food Partnership in 2018 (Table 4). However,some stated the main advocacy ask of getting thegovernment to establish targets for salt was notachieved within the intervention period. Many identi-fied a lack of coordinated advocacy actions and clearallocation of roles and responsibilities to have hin-dered intervention delivery. Uncertainty regardingwhich organisation was driving the strategy and whenorganisations were advocating as part of the Partner-ship or alone were suggested to have held up inter-vention delivery and led to sparse communication ofadvocacy activities between members, resulting in alack of coordinated action. The lack of allocation ofroles created uncertainty amongst individual membersas to who was responsible for carrying out the advo-cacy activities and who was responsible for makingsure they were executed (Table 2).

Contextual factors affecting intervention design, deliveryand outcomesStakeholders identified legal, political, social, environ-mental, technological and economic factors affectingintervention design, delivery and outcomes. Key themesand supporting quotes are provided in Table 4.The state and national food policy environments were

frequently discussed and perceived to shape the inter-vention design, hinder intervention delivery and preventthe Partnership achieving outcomes. At the state level,attempts to embed salt as a priority in current institu-tional nutrition policies (e.g. schools, hospitals and pub-lic sector workplaces) were thought by one stakeholderto be hindered by an inability to access relevant organi-sations to understand what programs were being

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 13 of 21

implemented and a lack of available data on implemen-tation or effectiveness. At the national level, slow pro-gress by the government’s Healthy Food Partnership insetting and implementing nutrient reformulation targetswas viewed as a barrier to effective action in two ways.Firstly, the announcement of the Healthy Food Partner-ship and its proposed program of work in 2016, requiredthe Partnership to adapt the Advocacy and Policy strat-egy to the shift in the policy environment. Subsequently,the slow progress in setting targets, including a lengthypublic consultation, impeded further advocacy attemptsfor policy change within the intervention timeframe.Secondly, the slow progress in target setting hindered In-dustry Engagement by the Partnership. Participants indi-cated that if targets had been in place there would havebeen an acceleration of industry reformulation activityand higher demand for the Partnership to support com-panies to reduce salt in food, and ultimately a greaterimpact on the food supply. Although one participantstated that the Partnership was able to put in place astrategy and “identify a program of work to engage in-dustry at the same time as waiting for the targets”(Member 2) (Table 4).The political and social climate was viewed as not con-

ducive to achieving Partnership aims. Many stakeholdersspoke about “a lack of recognition of salt as a populationhealth priority in the Australian context” (Member 13).Some speculated that a crowded and competitive nutri-tion space in the public domain hindered the Partner-ship from achieving political and social traction, withsugar and obesity being more “top of mind” issues. Somespoke of the link between consumer demand and gov-ernment and industry action. They suggested that a lackof public push for policy change continues to allow poli-ticians to ignore salt as a health priority and a lack ofpublic demand for healthy foods continues to allow in-dustry to produce processed foods high in salt. A fewstakeholders viewed a key role of the Partnership to becontinued advocacy on the importance of salt reductionto consumers, government and industry, building a busi-ness case for when policy window opens (Table 4).

DiscussionThis research, based on in-depth interviews with 14Partnership members and seven food industry stake-holders, has provided valuable insights into stakeholderperceptions on a partnership approach to implementinga salt reduction intervention. Use of CFIR enabled un-derstanding of factors affecting program implementationand how these factors were perceived to influence Part-nership effectiveness. This has added to our knowledgeon successes and challenges of working as a partnershipto deliver a multi-faceted public health intervention andprovided valuable lessons for future initiatives.

The PartnershipStakeholders viewed the establishment of the Partner-ship as essential for intervention planning, design anddevelopment, and an enabler for intervention delivery.Previous nutrition coalitions in Australia have tended tobe homogenous in knowledge and skills with mostmembers being from a nutrition background [20];whereas, effective health partnerships engage diversemembers who bring a wide range of skill sets and know-ledge [21, 22]. Consistent with this perspective, Partner-ship members viewed their diversity as a strength. ThePartnership brought together skills and expertise in di-verse areas such as communications, campaign manage-ment, public health interventions, disease preventionand treatment, advocacy and research. By working to-gether, an effective and diverse partnership can achievebetter outcomes than any individual or organisation canalone by sharing knowledge and skills, more-efficientlyutilising available resources and executing joint activities[23]. Interviewees described how the diversity within thegroup facilitated the cultivation of a positive learning cli-mate and they spoke positively about their ability to con-tribute to the strategy and their role in executing thePartnership’s plan. Thus, the effectiveness of interven-tion implementation was facilitated by the establishmentof a Partnership with diverse members who were able tocontribute their knowledge, skills and expertise, whichenabled effective collaborative action and interventionexecution.

Execution of the interventionThe interviewees felt the planned activities of each inter-vention arm were executed and the intended outputswere generated, though some recalled adaptations andadditions over the implementation period, and this wasviewed as an achievement. However, there were variedbeliefs regarding the translation of outputs to anticipatedoutcomes across the intervention arms [12]. This mayreflect the delay between outputs and outcomes com-monly seen in public health and policy interventions,but is also likely due to (1) internal factors, such as Part-nership communication networks, (2) contextual factors,such as the social and political context, and (3) interven-tion fidelity, which were highlighted in the interviews.

Communication networks within the PartnershipThe Partnership demonstrated some common featuresof effective collaborative frameworks, such as membercommitment, communication structures, openness,planning, a shared vision, and clear decision-makingprocesses [24]. For example, interviewees described clearbenefits associated with being a member of the Partner-ship, such as fulfilling organisational health promotiongoals, which were important in facilitating commitment

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 14 of 21

Table

4Con

textualfactorsaffectinginterven

tionde

sign

,deliveryandou

tcom

es

Con

textua

lfactors

Them

eQuo

tes

Policyen

vironm

ent

Health

yFood

Partne

rship’sslow

prog

ress

towards

settingdraftsodium

targets

Ifeelliketheadvocacy

activities

probablyha

vejustaddedastrong

supporterbase

around

things

that

werein-train.A

reform

ulationprogrammewas

beingdiscussed,beingdesig

ned,beingconsultedon

.Wecouldthen

putPartnershipre-

spon

sesin,the

technicalresponsethat

TheGeorgeInstitute

putin,orjustputtheweigh

tbehind

it–weneed

itfast,

weneed

itto

beareallycomprehensiveprogramme.(M

ember11:SP,I,R)

Ithink

we’ve

certainlyaffected

things

that

have

been

goingon

.The

HealthyFood

Partnershipwas

there,an

dwewere

ableto

offercommentarywheneverwewereasked,weofferedsupportan

dwewillcontinue...to

kind

ofpush

that

reform

ulationstorywith

government.So,w

hilst

wemight

notha

veseen

policychan

ge,w

e’ve

definitelycontinuedthe

conversationan

dputsupportbehind

it,extrasupportbehind

itwhenwe’ve

been

asked.(M

ember21:SP,I)

Itwas

challeng

ingto

getengagementbecausethefood

companies

werewaitingto

findoutwha

tthesaltreduction

targetsweregoingto

be…

theHeartFoun

dation,with

oursupport,man

ages

toidentifyaprogram

ofworkto

engage

industryat

thesametim

eas

waitingforthetargets.(M

ember2:SP,I,R)

Obviously,the

lagon

thetargetsfortheHealthyFood

Partnershipbeingsetha

sbeen

abitdisappointingan

dobviously

that’soutsidethecontrolofthe

group.Oncethosearein

placeyouwouldthinkthat

that

may

heighten

the

activity

ofthefood

industryan

dman

ufacturersto

reform

ulatean

dhelpto

reachthat

outcom

ebutagainthat’sstilla

whileaw

ay.(Mem

ber1:SP,R)

Alignm

entwith

statenu

trition

policies/

strategies

(e.g.H

ealth

yCho

ices)

Idon

’tthink[salt]isnecessarily

embedded

instrategy

at[state]governmentlevel...

there’safewindividualsthat

have

reallydriven

itan

dcham

pion

editwhich

hasbeen

fantastic,but

whetherthat’sem

bedded

inthestrategy

ifthey

were

toleave,I’m

stillno

tsure.(Mem

ber7:SP)

Inrelationto

thestate-widepolicieson

food

ininstitutiona

lsettings,it’s

been

achalleng

eto

impacton

thoseprocesses

andalso

toknow

whetheror

notthosepoliciesarebeingimplem

entedan

dwhetheror

notthey

arealso

incorporating

saltin

thosepolicies...wedidn

’tha

vean

ydirectaccessto

thoseinstitutiona

lsettings.(Mem

ber2:SP,I,R)

Politicalclim

ate

Saltisno

tacurren

tgo

vernmen

tpriority

There’sbeen

somegood

workbutas

awho

leprobablyha

sn’tquite

hadtheimpactat

thestatean

dfederallevelas

wewouldha

vewan

ted,an

dthat’sno

tnecessarily

becauseof

thefaultof

anyof

thepartners,it’s

partlybecauseof

the

politicalconversations

andagendasoutwheresaltisan

dyoucan’tmakean

issue

popularwith

politicians

ifthey

don’twan

titto

bean

dthere’sno

tapublicpush...so

thepoliticians

have

been

ableto

kind

ofigno

reit.(M

ember10:

SP)

Weobviously

wan

tedpolicychan

ge.W

eha

veha

dagovernmentin

...it’sno

tthehigh

estthingon

theirprioritylist.

(Mem

ber21:SP,I)

Lack

ofpo

licywindo

win

theinterven

tion

timepe

riod

Itwouldha

veto

dowith

thefactthat

wha

twe’retrying

todo

was

trying

tomoveamountainan

dthat’savery

difficultthingto

dowhenyou’rejustagroupof

mountaineers...The

reality

isit’sstillapriority,itneedsto

beapriority

andlikein

tobaccocontroloran

yotherpublichealth

area,oftenitcantake

twentyor

thirtyyearsfortheadvocacy

reallyto

reachacriticalm

assan

dthen

findasympatheticministeror

asympatheticgovernmentor

asympathetic

commun

ityan

dthetim

ingisright

andsuddenlyyougetan

opportun

ity…

Alotof

successof

publichealth

hastim

eto

build

andalong

timeto

makethebusin

esscase,and

then

thepolicywindowopens,an

dthat’stheopportun

itythere.(M

ember13:SP)

Sometimes

you’ve

justgotto

gowith

wha

tthepoliticalclimate’sgoingto

letyoudo,and

both

with

industryan

dpoliticians

andsometimes

it’sno

tyour

dayor

year

oryour

decade

andyoucan’tchan

geeverything

unfortun

ately.

(Mem

ber10:SP)

Socialfactors

Crowde

dnu

trition

space

Thereality

isthat

thebroadcommun

itydebate

ontheseissuesha

veusually

been

dominated

byissuesof

asugartax

orhealth

levy

onsugarydrinks

andtheroleof

sugaras

adriverof

theobesity

epidem

icin

recent

years,an

dIthink

tothat

extent

thePartnershipan

dthesaltfocuswas

alwaysgoingto

struggleto

getcutthrough,butitdoesn’tmean

that

it’safailure,itjustmeans

that

it’sacompetitiveenvironm

entforcommun

ityengagementon

food

andIthink

there’sbeen

somegood

progressmadeby

thePartnershipin

generalcom

mun

ityaw

areness,engaging

with

the

industry,eng

agem

entwith

policymakersas

wellbut

that’sagainstthebackdrop

ofbroadlyalack

ofrecogn

ition

ofsaltas

apopulationhealth

priorityin

theAu

straliancontext.(M

ember13:SP)

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 15 of 21

Table

4Con

textualfactorsaffectinginterven

tionde

sign

,deliveryandou

tcom

es(Con

tinued)

Con

textua

lfactors

Them

eQuo

tes

Sugarwas

muchmoreof

atopof

mindissue

forman

y,particularlyconsum

ers,butalso

forindustryan

deven

governmentto

somedegree,orthepolitical

endof

thespectrum

.(Mem

ber7:SP)

Perceivedvalueof

asing

lenu

trient

approach

Not

everyone

inVictoria,and

inVictoriahealthyeatingsector,w

ouldsupportasaltreductionapproach...nu

trient-

specificprojectsareno

treallywha

tthey’re

atthemom

ent,prioritising

.(Mem

ber11:SP,I,R)

I’vealwaysbeen

averystrong

internaladvocate

inthefood

andnu

trition

spacean

dparticularlyaround

theneed

for

afocuson

saltreductionan

ditha

sbeen

oneissue

that

reallywas

becomingalittle

problematicwas

thisbroader

issue

that’srunby

nutrition

ists,in

general,dietitian

sIsho

uldsay,aboutthefactthat

weshouldn’tfocuson

single

nutrients.(M

ember8:SP)

Environm

entalfactors

Thebroade

rfood

environm

ent

Everyone

inAu

stralia

gets70%

oftheirsaltor

80%

oftheirsaltfro

mpackaged

foodsan

drestaurant

foods.(M

ember4:

SP,R)

Techno

logicalfactors

Techno

logy

asabarrierto

reform

ulation

Barriersaretaste,techno

logy

isto

doit...consum

eracceptan

ce,qualitybecausesaltprovides

atechnicalrole,no

tjust

boostingflavours,though

that’sdefinitelyafactor.(Industry17)

Therearethings

that

wecando

andtherearethings

that

wecan’tdo...also

theimplications

ofreform

ulation,wha

tdo

wereplaceitwith,ifa

nything,an

disthat

moreor

lessha

rmfultha

ntheoriginal.From

ourperspectiveifweask

thefood

industryto

reform

ulatein

ashortperiodof

time,they

canon

lyusethetechno

logy

that’savailableto

them

,butyouknow

doesn’tgive

them

timeto

look

forotheroptions

andyoumay

getafood

supplyflooded

with

things

that

youdon’twan

t.An

dthat’swha

tweareverymindfulof

andwe’reveryconcerned...wetryno

tto

putotherthings

inthat

may

notbe

necessarily

abetteroption.(Industry18)

Wealso

have

jointresearch

programs...totryan

dusedifferent

techno

logies

sowecangetthesodium

downbutstill

have

agood

taste.(Industry19)

Other

factorsaffectingthefood

indu

stry’sability

toreform

ulate

Functio

nalroleof

saltin

food

Inman

ycases,especiallyin

extrusionan

din

othertypesof

crisp

ing,there’son

lyso

muchyoucantake

outbeforethe

productno

long

erresembles

itsform

.(Industry18)

Insomeotherindustries...you

need

saltforafunctiona

lbenefit,an

dalso

insomeof

theproductswedo

likecrackers

wheresodium

bicarbon

ateisused

asaraising

agent,it’ssometimes

veryha

rdthen

togetproductslowin

acid

sodium

.Sodium

canbe

used

eitheralon

eor

incombina

tionwith

otherthings

likesodium

bicarbon

ate,likea

functiona

lbase,likearaising

agent.(Industry19)

Thereisasw

eetspot

where

man

ufacturerscantake

outsm

allamountsof

sodium

withouttoonegativelyaffecting

tastean

dfunction,an

dIthink

respon

siblefood

producersareworking

towards

achievingthat.Inorderto

gofurther,

which

isreallybecomingcreativewith

ingredientsan

dflavoursan

dso

onto

reduce

thesodium

,but

stillmaintain

terrifically

tastyproductsforconsum

ers...Iftheydon’tha

vethefood

scienceproductdevelopm

entteam

sto

workwith

ingredientsan

dflavoursan

dso

onto

produceoutstand

ingproductswith

taste,with

lowsodium

...wha

tthey

willneed

ispossiblyassistanceor

directionwith,and

exposureto

askillsetan

dinform

ationthat

willallowthem

toexplore

productdevelopm

entan

dreform

ulationat

lowsodium

target

levels.

(Industry14)

Feasibility

ofremovingsaltfro

mfood

Inthecompany’sthat

Ihaveworkedforor

consultedto,theyha

veliterallyjustremoved

thesodium

becausetheir

productsaren’tverycomplicated,theyha

ven’tha

dto

look

foralternatives.(Industry16)

Thethingwith

saltisthat

it’seasierto

removean

dreplacesaltthan

itisto

removean

dreplacesugar.So,itis

something

that

youcando

bystealth

andtrainthepalate

slowlyto

getused

to.A

ndbecausethereisverylittle

bulkin

saltit’sjusteasierto

take

outthan

someof

theotheringredients.(Industry18)

Skillsandexpe

rtiseto

reform

ulate

Und

erstan

ding

that

some,especiallythesm

allerman

ufacturers,m

ight

need

someskilled

food

sciencean

dproduct

developm

enthelpin

reform

ulatinglowersodium

,but

stillmaintaining

amazingmouth

feelan

dtaste.(Industry14)

Companies

nowem

ployingdietitian

san

dregistered

nutrition

ists.That’simportan

tthat

they

acknow

ledgethat

they

need

that

levelofexpertisean

dsciencewithin

theircompany

tohelpguidethem

,tohelpguidetheirleadershipteam

towha

tthey

shouldbe

doingfro

mahealth

perspective.(Industry17)

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 16 of 21

Table

4Con

textualfactorsaffectinginterven

tionde

sign

,deliveryandou

tcom

es(Con

tinued)

Con

textua

lfactors

Them

eQuo

tes

Thebigplayersaretheon

eswho

aredoingalotin

that

spacebecausethey’vegottheresource,likewe’ve

gota

dedicatednu

trition

team

,adedicatedregulatoryteam

,asustaina

bilitynu

trition

team

,we’ve

gottheresources.

Whereas

Ikno

wthesm

allerbusin

essesdon’tan

dthey

need

toseewha

twe’redoing,an

dthey

need

tolearnfro

mus.

(Industry17)

There’sahu

gegapin

resourcesan

dhu

gegaps

incapabilities.Even

thelargeman

ufacturerswho

wena

ivelythough

tha

dthecapabilitiesdo

notha

vewell-resourced

nutrition

units

within

theirorganisations.(Mem

ber6:I)

Costof

reform

ulating

Ithink

costisamassiveissue.Buyingsaltreplacem

entsismoreexpensivethan

puttingsaltin

andit’sjustho

wwilling

peopleareto

beableto

take

ahiton

grow

thmargins,and

ifyoudo

itslo

wlyby

stealth

then

you’rein

abetter

position.(Industry18)

Everysin

glereform

ulationthat

youdo

requiresaproject,requiresresources,requirestrialtime,requirestrialm

anaging,

requirespackagechan

ges,it’saveryexpensiveexercise.(Industry18)

Saltisacheapingredient

anditcanbe

that

aswereduce

salt,thecostsgo

up.(Industry2)

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 17 of 21

to the Partnership and collaborative working [23, 25].Stakeholders also described some qualities that couldhave been improved. Interviewees discussed communi-cation challenges encountered between members work-ing at the strategic level and program implementers.Despite quarterly strategic meetings, there was a lack oflack of clarity around the overall vision of the interven-tion, with stakeholders from these distinct groups hold-ing different views of action area aims and objectives,and individual’s roles and responsibilities [22]. Hunterand Perkins propose that a lack of connection betweenmembers at the strategic-level and implementation-levelis a barrier to effective partnership working due to a lackof transfer of knowledge, skills and expertise betweengroups [26]. They found that strategic partners were fo-cused on achieving the strategic goals and targets, whilethose involved in intervention delivery operated in amore organic way, distinct from the objectives set at thestrategic level [26]. This was evident within the currentPartnership. Strategic-level members mostly spoke aboutachieving strategic aims and objectives that were deter-mined at the outset, some of which were viewed as tooambitious; while program implementers seemed to havevaried understandings of the strategic goals and weremore focused on meeting organisational and individualcontractual obligations (e.g. key performance indicators).The disparity between individuals’ perspectives on Part-nership goals highlights a crucial gap between the expec-tations of the strategic partners and the ability of theprogram implementers to respond.Furthermore, program implementers and strategic

partnership members held different perspectives on eachother’s roles. For example, strategic leaders viewed theirrole as central to the intervention design and delivery,while program implementers thought of the role of stra-tegic partners as less important. Program implementersshared details about their decision-making processes,which tended to be made at the implementation teamlevel and independently of the strategic goals. These de-cisions, combined with staff turnover and poor handoverwithin the implementation team, resulted in subsequentchanges to the intervention plan and ultimately changedthe ways in which the intervention team thought aboutand executed the overall Partnership plan. Given sharedunderstanding of clear aims and objectives, and delega-tion of roles and responsibilities, are factors that facili-tate Partnership functioning [25, 26], the differences inindividual perspectives that were highlighted in the in-terviews, may have reduced the effectiveness of the Part-nership. Without having a solid understanding of theanticipated outcomes, the intervention team were work-ing towards generating the outputs they were contractedto produce rather than striving to achieve the plannedstrategic outcomes through the outputs [26, 27]. This

highlights the need for partnerships to ensure relation-ships and the transfer of knowledge and skills are main-tained between those at strategic levels and thoseresponsible for intervention delivery [26].

Contextual factorsThe importance of planning and implementing an ap-propriate intervention for the context was highlighted,as well as planning within the constraints of a chosenintervention approach, such as within the limitations ofa voluntary state-based program. This salt reductionproject utilised a partnership approach to execute afour-armed salt reduction intervention plus monitoringand evaluation. Around the world, countries are employ-ing multi-dimensional salt reduction strategies to reduceexcessive population-level salt intakes and the associateddisease burden [28]. The Partnership approach com-prised evidence-based salt reduction strategies that havebeen previously identified, including: Measuring andmonitoring salt use and knowledge, attitudes and behav-iours of Victorians and salt levels in the Australian foodsupply, engaging the food industry to reformulate prod-ucts, influencing the establishment of sodium targets forfoods and front-of-pack labelling in Australia, consumereducation to empower individuals to change their saltuse behaviours and supporting the creation of healthyfood environments in public institutions [29]. Simplyimplementing a multi-faceted evidence-based interven-tion is not enough to achieve public health outcomes ina complex real-world setting where contextual factorsare at play [27]. Tseng suggests it is not just about whatworks, but “what works for whom, under what condi-tions and at what cost” [30]. Two key challenges wereidentified: executing a salt reduction strategy within oneAustralian state from a state-based perspective and lackof available policy levers stemming from the delegationof power between state/territory governments and theCommonwealth government in Australia. Legislative andregulatory responsibilities for health policy and healthcare are split between state and Federal governments[31], which has resulted in challenges executing publichealth responses both now and in the past [32]. The de-livery of the Policy and Advocacy arm was hindered bycomplexities around this division of power, where somedesired activities needed to be executed at the federal-level (e.g. advocating for sodium targets) and others atthe state-level (e.g. integrating salt reduction into institu-tional nutrition policies). Partnership resources for thisintervention arm were divided and stretched betweenthe two approaches, and ultimately this resulted in agreater focus on Federal government advocacy and in-attention to the state-based goal. Notably, at the Federallevel, the few government strategies that have aimed toimprove population diets have been voluntary and are

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 18 of 21

unlikely to have much impact [33, 34], even though sys-tematic reviews and modelling studies suggestmandatory or legislative approaches may be more effect-ive [35, 36]. Regardless of this, through its program, thePartnership was able to engage with the food industry,including successfully building positive relationships andfacilitating reformulation [7], in the absence of regula-tion. This example demonstrates that the Partnershipapproach and strategy could be a useful model for theGovernment to adopt. A key lesson from the Partnershipwas the need to understand the context of the interven-tion – the legal and political mechanisms that influencepublic health agendas – in order to plan an appropriateintervention that fits within the context and engageswith these mechanisms, prior to commencing theintervention.Further to this, the political and social climates were

viewed as barriers to achieving planned outcomes; par-ticularly policy change, given the lack of public and pol-itical push for salt reduction. To achieve nutrition policychange, gaining public support, and demonstrating it topolicymakers, is crucial for generating political will [37].As a newly-formed coalition, the Partnership firstlyneeded to gain credibility and trust from the public [37].This is often hard for nutrition coalitions to achieve asthere is a perception from the public that the nutritionmessage is constantly changing and there is lack of con-sensus among nutrition professionals [37]. Members be-lieved a focus on salt reduction alone made it difficult tocompete with more topical issues such as sugar andobesity. The public nutrition space was described as“crowded” by mixed messaging from various specialinterest groups making it difficult for the intervention’smessages to stand out and for the Partnership to build acredible reputation, which likely limited its influence onpublic will. Interestingly, even with substantial supportand longer-term advocacy efforts, public health and nu-trition professionals have not managed to influence thegovernment to implement a national nutrition policy[38]. It is important to note, that the Partnership wasable to influence the government’s sodium target settingprocess, and subsequent to the intervention (May 2020),the government announced 27 sodium reduction targets[39]. This suggests public health and nutrition coalitionscan influence policies already on policymakers’ agendasbut illustrates the challenge in generating the public andpolitical will needed to get policies onto decision-makers’ agendas.

Fidelity of the interventionA monitoring and evaluation process was actualised to en-sure intervention fidelity and generate the evidenceneeded to support decision-making to optimise the effect-iveness of the Partnership intervention throughout the

implementation period. Our analysis revealed that theprogram implementers were able to incorporate feedbackand adapt components of the intervention plan in re-sponse to Partnership monitoring and evaluation efforts[14]. For example, the need for a more strategic approachthat would have greater impact on the food supply by tar-geting both small and medium businesses as well as largermanufacturers through a variety of different engagementmethods was identified in an interim assessment [14]. Ef-fective partnerships have clear strategic plans, which aresupported by robust monitoring and evaluation proce-dures, to determine how the partnership is performingand how intervention delivery is going throughout theintervention to establish if any modifications are needed[26]. Planned and unplanned adaptations are often madeprior to interventions (e.g. for a specific audience or con-text) and during interventions (e.g. to optimise effective-ness) [40]. Adaptations to intervention plans can have apositive or negative impact on the intervention’s effective-ness [41, 42]. Following the intervention, stakeholdersviewed the food industry strategy adaptation as one of thekey successes and a facilitator in achieving Partnershipoutcomes. However, not all recommended changes wereexecuted by the implementation team, which suggeststhere were missed opportunities to increase the impact ofthe Partnership intervention. For instance, in 2017 and2019 stakeholders identified a need for stronger, strategicconsumer-messaging that conveyed the serious healthrisks of high salt [14]; however, perceptions on thestrength of awareness-raising messages remained un-changed throughout the interventions. Program imple-menters did not comment on why such feedback was notincorporated; however, this was potentially due to achange in key personnel. They did speak about an unsuc-cessful attempt to shift consumers along the Transtheore-tical Model stages of change [43] through the threecampaign waves. While the Transtheoretical Model canbe applied to public health interventions to try to acceler-ate the rate of behaviour change, only a small proportionof the population will be willing to contemplate making achange and an even smaller proportion will be ready totake action within the intervention timeframe, leading tovaried successes in changing dietary behaviour in popula-tions [44]. This is an example of prioritising interventionfidelity over adaptation and optimisation, and raises ques-tions about whether the Partnership could have made fur-ther progress towards its consumer awareness goals if theinterim feedback had been applied.

Strengths and limitationsThe strengths of this study include that the interviewtool had been previously used for the 2017 interviews[14] and was built on survey instruments used in previ-ous studies [15, 16]. The semi-structured approach

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 19 of 21

allowed interviewers to ask tailored questions to eachinterviewee given their involvement in the Partnershipand ask probing questions to gain further informationbased on participant responses. The qualitative analysiswas guided by the CFIR, which facilitated the identifica-tion and understanding of the internal and contextualfactors that influenced intervention implementation andeffectiveness [17]. Some limitations are noted. Althoughall Partnership members were invited to participate andthere was a variety of strategic, implementation and re-search members, only 14/24 (58%) Partnership membersagreed to participate and most interviewees were fromorganisations involved in implementation or evaluation.All food manufacturers engaged by the implementationteam were also invited to participate, however only 7/19(37%) stakeholders agreed to participate, and these inter-viewees were from larger companies and their experi-ences may not be representative of all companiesengaged.

ConclusionsThe establishment of a Partnership with diverse skillsand experience facilitated collaborative action and inter-vention delivery. Monitoring and evaluating implemen-tation informed strategy adaptations which allowedoptimisation of Partnership strategy. Future partnershipsshould consider the importance of developing strongcommunication networks, particularly between strategicand implementation-levels, interventions that fit thecontext and utilise available contextual mechanisms, andthe balance between intervention adaptation and main-taining intervention fidelity.

AbbreviationsDALY: Disability-adjusted life year; TGI: The George Institute for Global Health;IPAN: Institute for Physical Activity and Nutrition; SGSA: State government orstatutory agency; CFIR: Consolidated framework for implementation research

AcknowledgementsThe authors wish to thank the interviewees for participating in the study.

Authors’ contributionsE.R., B.M., K.T., J.R. and J.W. contributed to the design of the study. E.R. andW-K.C. conducted the interviews. E.R. analysed and interpreted the data anddrafted the manuscript. All authors, E.R., W-K.C., B.M., J.R., K-A.J., K.C., K.T. andJ.W., contributed to the final manuscript by providing input into the inter-pretation of the data, reviewing, and editing the manuscript. All authors haveread and approved the final manuscript.

FundingThis study was funded by a National Health and Medical Research CouncilProject Partnership Grant (APP1111457) with additional funding from theVictorian Health Promotion Foundation. ER is supported by a University ofNew South Wales University Postgraduate Award (UPA) and George InstituteTop-Up Scholarship. BM is supported by a UNSW Scientia PhD scholarship.KT is supported by an Early Career Fellowship from the National Health andMedical Research Council of Australia (APP1161597) and a Postdoctoral Fel-lowship from the National Heart Foundation of Australia (Award ID 102140).JW is supported by a National Heart Foundation Future Leaders Fellowship II(#102039), a NHMRC CRE on food policy interventions to reduce salt(#1117300) and NHMRC project grants (#1052555 and #1111457).

Availability of data and materialsThe datasets used and/or analysed during the current study are availablefrom the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participateThis study was approved by the University of Sydney Human Ethics ResearchCommittee (2016/770) and performed in accordance with the Declaration ofHelsinki. Written informed consent was obtained from all participants priorto the interview.

Consent for publicationNot applicable.

Competing interestsJ.W. is Director of the World Health Organization (WHO) Collaborating Centreon Population Salt Reduction. VicHealth initiated and funded the VictorianSalt Reduction Partnership, J.R. and K.C. were not involved in interviewingstakeholders, analysing the data or reporting the results. All other authorsdeclare that they have no competing interests.

Author details1The George Institute for Global Health, The University of New South Wales,Sydney, NSW 2006, Australia. 2Victorian Health Promotion Foundation,Melbourne, VIC 3053, Australia. 3National Heart Foundation of Australia,Melbourne, VIC 3008, Australia.

Received: 12 November 2020 Accepted: 18 February 2021

References1. World Health Organization. ‘Best buys’ and other recommended

interventions for the prevention and control of noncommunicable diseases.Updated (2017) Appendix 3 of the global action plan for the preventionand control of noncommunicable diseases 2013–2020. Geneva: WorldHealth Organization; 2017.

2. World Health Organization. Guideline: sodium intake for adults and children.Geneva: World Health Organization; 2012.

3. Powles J, Fahimi S, Micha R, Khatibzadeh S, Shi P, Ezzati M, et al. Global,regional and national sodium intakes in 1990 and 2010: a systematicanalysis of 24 h urinary sodium excretion and dietary surveys worldwide.BMJ Open. 2013;3(12):e003733.

4. Institute for Health Metrics and Evaluation (IHME). GBD Compare DataVisualization. Seattle: IHME, University of Washington; 2017. Available from:https://vizhub.healthdata.org/gbd-compare/

5. World Health Organization. Global status report on noncommunicablediseases 2014. Geneva: World Health Organization; 2014. Available from:https://apps.who.int/iris/bitstream/handle/10665/148114/9789241564854_eng.pdf;jsessionid=98B7854EE6D6173E4848891BECFF3D27?sequence=1

6. Webster J, Trieu K, Dunford E, Nowson C, Jolly K-A, Greenland R, et al. Saltreduction in Australia: from advocacy to action. Cardiovasc Diagn Ther.2015;5:207–18.

7. Rosewarne E, Trieu K, Farrand C, Reimers J, Potter J, Davidson C, et al.Unpack the salt: an assessment of the Victorian salt reduction Partnership’smedia advocacy activities to highlight the salt content of different foods.Nutr J. 2020;19(1):102.

8. Bolton KA, Webster J, Dunford EK, Jan S, Woodward M, Bolam B, et al.Sources of dietary sodium and implications for a state-wide salt reductionprogram in Victoria, Australia. Br J Nutr. 2020;123(10):1165–75.

9. Grimes CA, Riddell LJ, Campbell KJ, Beckford K, Baxter JR, He FJ, et al.Dietary intake and sources of sodium and potassium among Australianschoolchildren: results from the cross-sectional salt and other nutrients inchildren (SONIC) study. J BMJ open. 2017;7(10):e016639.

10. Victorian Health Promotion Foundation (VicHealth). Salt Reduction inVictoria. Melbourne: VicHealth; 2017. Available from: https://www.vichealth.vic.gov.au/programs-and-projects/salt-reduction

11. Christoforou A, Trieu K, Land M-A, Bolam B, Webster J. State-level andcommunity-level salt reduction initiatives: a systematic review of globalprogrammes and their impact. J Epidemiol Community Health. 2016;70(11):1140–50.

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 20 of 21

12. Trieu K, Jan S, Woodward M, Grimes C, Bolam B, Nowson C, et al. Protocolfor the process evaluation of a complex, statewide intervention to reducesalt intake in Victoria, Australia. Nutrients. 2018;10(8):998.

13. Victorian Health Promotion Foundation (VicHealth). State of Salt: The case forsalt reduction in Victoria. Melbourne: VicHealth; 2015. Available from: https://www.vichealth.vic.gov.au/media-and-resources/publications/state-of-salt

14. McKenzie B, Trieu K, Grimes CA, Reimers J, Webster J. Understandingbarriers and enablers to state action on salt: analysis of stakeholderperceptions of the VicHealth salt reduction partnership. Nutrients. 2019;11(1):184.

15. Trieu K, Webster J, Jan S, Hope S, Naseri T, Ieremia M, et al. Processevaluation of Samoa’s national salt reduction strategy (MASIMA): whatinterventions can be successfully replicated in lower-income countries?Implement Sci. 2018;13(1):107.

16. Webster J, Pillay A, Suku A, Gohil P, Santos J, Schultz J, et al. Processevaluation and costing of a multifaceted population-wide intervention toreduce salt consumption in Fiji. Nutrients. 2018;10(2):155.

17. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC.Fostering implementation of health services research findings into practice:a consolidated framework for advancing implementation science.Implement Sci. 2009;4(1):50.

18. Bradley EH, Curry LA, Devers KJ. Qualitative data analysis for health servicesresearch: developing taxonomy, themes, and theory. Health Serv Res. 2007;42(4):1758–72.

19. Food & Drink Business. Australia's top 100 Food & Drink Companies 2019.Food & Drink Business November–December 2019; 2019. p. 29–57.

20. Cullerton K, Donnet T, Lee A, Gallegos D. Exploring power and influence innutrition policy in Australia. Obes Rev. 2016;17(12):1218–25.

21. Willis C, Greene J, Abramowicz A, Riley B. Strengthening the evidence andaction on multi-sectoral partnerships in public health: an action researchinitiative. Chronic Dis Injuries Canada. 2016;36(6):101.

22. Baker EA, Wilkerson R, Brennan LK. Identifying the Role of CommunityPartnerships in Creating Change to Support Active Living. Am J Prev Med.2012;43(5, Supplement 4):S290–S9.

23. Butterfoss FD, Kegler M. The community coalition action theory. In:DiClemente RJ, RAC, editors. Emerging theories in health promotionpractice and research. 2nd ed. San Francisco: Jossey-Bass; 2009. p. 237–76.

24. Willis C, Greene J, Riley B. Understanding and improving multi-Sectoralpartnerships for chronic disease prevention: blending conceptual andpractical insights. Evid Policy. 2017;13(4):623.

25. Taylor-Robinson DC, Lloyd-Williams F, Orton L, Moonan M, O'Flaherty M,Capewell S. Barriers to partnership working in public health: a qualitativestudy. PLoS One. 2012;7(1):e29536.

26. Hunter D, Perkins N. Partnership working in public health: the implicationsfor governance of a systems approach. J Health Serv Res Policy. 2012;17(2_suppl):45–52.

27. Wandersman A, Alia K, Cook BS, Hsu LL, Ramaswamy R. Evidence-basedinterventions are necessary but not sufficient for achieving outcomesin each setting in a complex world: empowerment evaluation, gettingto outcomes, and demonstrating accountability. Am J Eval. 2016;37(4):544–61.

28. Trieu K, Neal B, Hawkes C, Dunford E, Campbell N, Rodriguez-Fernandez R,et al. Salt reduction initiatives around the world - a systematic review ofProgress towards the global target. Plos One. 2015;10(7):e0130247.

29. World Health Organization. SHAKE the Salt Habit. The SHAKE TechnicalPackage for Salt Reduction. Geneva: World Health Organization; 2016.Available from: https://apps.who.int/iris/bitstream/handle/10665/250135/9789241511346-eng.pdf;jsessionid=8FB361AB384DFE9F6F6732A2A3F4FBCE?sequence=1

30. Tseng V. Evidence at the crossroads pt. 11: The next generation ofevidence-based policy. New York: William T Grant Foundation; 2016.

31. Parkin A, Summers J, Woodward DF. Government, politics, power and policyin Australia. Melbourne: Pearson Australia Group; 2006.

32. Moloney K, Moloney S. Australian quarantine policy: from centralization tocoordination with mid-pandemic COVID-19 shifts. Public Adm Rev. 2020;80(4):671–82.

33. Jones A, Shahid M, Neal B. Uptake of Australia’s health star rating system.Nutrients. 2018;10(8):997.

34. Rosewarne E, Huang L, Farrand C, Coyle D, Pettigrew S, Jones A, et al.Assessing the healthy food Partnership’s proposed nutrient reformulationtargets for foods and beverages in Australia. Nutrients. 2020;12(5):1346.

35. Hyseni L, Elliot-Green A, Lloyd-Williams F, Kypridemos C, O'Flaherty M,McGill R, et al. Systematic review of dietary salt reduction policies: evidencefor an effectiveness hierarchy? Plos One. 2017;12(5):e0177535.

36. Nghiem N, Blakely T, Cobiac LJ, Pearson AL, Wilson N. Health and economicimpacts of eight different dietary salt reduction interventions. Plos One.2015;10(4):e0123915.

37. Cullerton K, Donnet T, Lee A, Gallegos D. Effective advocacy strategies forinfluencing government nutrition policy: a conceptual model. Int J BehavNutr Phys Act. 2018;15(1):83.

38. Public Health Association of Australia (PHAA). National Nutrition Policy.Policy Position Statement. Canberra: PHAA; 2018. Available from: https://www.phaa.net.au/documents/item/3287

39. Healthy Food Partnership. Reformulation Targets. Canberra: Commonwealthof Australia; 2020. Available from: https://www1.health.gov.au/internet/main/publishing.nsf/Content/reformulation-targets

40. Escoffery C, Lebow-Skelley E, Haardoerfer R, Boing E, Udelson H, Wood R,et al. A systematic review of adaptations of evidence-based public healthinterventions globally. Implement Sci. 2018;13(1):125.

41. Bopp M, Saunders RP, Lattimore D. The tug-of-war: fidelity versusadaptation throughout the health promotion program life cycle. J PrimPrev. 2013;34(3):193–207.

42. Pérez D, Van der Stuyft P, del Carmen ZM, Castro M, Lefèvre P. A modifiedtheoretical framework to assess implementation fidelity of adaptive publichealth interventions. Implement Sci. 2015;11(1):91.

43. Prochaska JO, DiClemente CC. The transtheoretical approach. HandbookPsychother Integration. 2005;2:147–71.

44. Greene GW, Rossi SR, Rossi JS, Velicer WF. Dietary applications of the stagesof change model. J Acad Nutr Diet. 1999;99(6):673.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Rosewarne et al. BMC Nutrition (2021) 7:12 Page 21 of 21