Social marketing interventions for the prevention and control of neglected tropical diseases: A systematic review

Background Social marketing is an approach to behavior change that contributes to disease prevention and control. This study aimed to understand how social marketing interventions have addressed neglected tropical diseases (NTDs). It examined the characteristics, breadth of coverage, and outcomes of social marketing interventions focused on the prevention and control of these diseases. Methodology/Principal findings Studies published in any language between January 1971 and April 2017, targeting at least one of the 17 NTDs prioritized in the World Health Organization (WHO) NTD Roadmap were considered. Included studies had interventions that applied both, at least one core social marketing concept, “social behavioral influence”, and one social marketing technique, “integrated intervention mix”, described in the Hierarchical Model of Social Marketing. This review is registered with PROSPERO CRD42017063858. Twenty interventions, addressing eight NTDs, met the inclusion criteria. They focused on behaviors related to four of the five WHO public health strategies for NTDs. Most interventions incorporated the concepts “relationship building” and “public / people orientation focus”, and the technique “insight-driven segmentation”. All the interventions reported changing behavioral determinants such as knowledge, 19 reported behavior change, and four influenced health outcomes. Conclusion/Significance Evidence from this study shows that social marketing has been successfully used to address behaviors related to most of the five public health strategic interventions for NTDs recommended by the WHO. It is suggested that social marketing interventions for the prevention and control of NTDs be grounded on an understanding of the audience and adapted to the contexts intervened. Building stakeholder relationships as early as possible, and involving the publics could help in reaching NTD outcomes. Elements of the intervention mix should be integrated and mutually supportive. Incorporating health education and capacity building, as well as being culturally appropriate, is also relevant. It is recommended that ongoing discussions to formulate the targets and milestones of the new global Roadmap for NTDs integrate social marketing as an approach to overcome these diseases.


Introduction
The World Health Organization (WHO) prioritizes a group of neglected tropical diseases (NTDs) [1,2], which affect approximately one billion people around the world [3]. These are also known as "diseases of poverty" [4] or diseases of neglected populations [5]. These diseases affect the most vulnerable populations everywhere, not exclusively those living in lowerincome economies [4,6]. NTDs pose a challenge for the health, social, and economic development of countries [5,7]. These challenges are exacerbated by changes in the global landscape, such as climate change, conflicts, migration, and urbanization processes [6,[8][9][10].
Varied strategies exist to address NTDs. The WHO prioritizes five public health strategic interventions that include promoting behavior change [4,11]. The priority strategies are: 1) preventive chemotherapy and transmission control (PCT), 2) innovative and intensified disease management (IDM), 3) vector ecology and management (VEM), 4) veterinary public health measures, and 5) the provision of safe water, sanitation, and hygiene (WASH). Efforts to support behavior change to prevent and control NTDs are needed [4,12], and social marketing could contribute to such efforts.
"Social marketing seeks to develop and integrate marketing concepts with other approaches to influence behaviors that benefit individuals and communities for the greater social good. Social marketing practice is guided by ethical principles. It seeks to integrate research, best

Study protocol
This systematic review was registered with PROSPERO (CRD42017063858) [31]. The protocol, published elsewhere [32], followed the guidelines of the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) [33].

Eligibility criteria
Peer-reviewed primary studies published between January 1971, the year the social marketing concept was coined [14], and April 2017, in any language, were included if they targeted the prevention or control of any of the 17 NTDs prioritized in the WHO NTD Roadmap [34]. No restrictions were made in the countries to include because NTDs affect the most vulnerable in low income and wealthier economies. No limits were set on the diseases as the study intended to identify the variety of priority diseases addressed by social marketing interventions.
Studies were retrieved if the interventions were labeled as social marketing or used social marketing terminology in their title or abstract, such as "marketing mix", "segmentation", "exchange", "social offering", and "behavior influence". Studies that applied at least one concept, "social behavioral influence", and one technique, "integrated intervention mix" were included in the review. These two criteria were established because they are common in social marketing interventions [17,24,[35][36][37][38], and because interventions dated before 2015, the year the Hierarchical Model of Social Marketing was published, might explicitly state that they adhered to these criteria. Moreover, to be included, the interventions had to report their intended or actual influence on behavioral determinants (e.g., attitudes, awareness, knowledge, infrastructure or policy change), behavior change, or health outcomes. Reviews and systematic reviews were excluded.

Information sources and search strategies
A literature search was conducted in PubMed, EbscoHost, ProQuest, Web of Science, Global Index Medicus, and Virtual Health Library Regional Portal. The search strategies are available in the protocol [32], at PROSPERO (CRD42017063858) [31], and in the supporting information (see S1 File). Reference lists of the selected studies were searched to identify additional relevant articles. When full-text articles were not available, corresponding authors were contacted. Some authors provided further publications describing the interventions and papers about interventions that had not been found through the literature search.

Modifications to the Hierarchical Model of Social Marketing
The Hierarchical Model of Social Marketing was used as a framework for analysis for several reasons. It is built on other social marketing frameworks, especially on the social marketing benchmark criteria; it is the most recent framework to describe components of social marketing interventions; and it uncovers elements of social marketing interventions that are not explicitly mentioned by other frameworks (e.g. relationship building, co-creation). Modifications were made to the labels of the items in the framework, as well as to some descriptions. The name of the concept "citizen/customer/civic society orientation focus" was changed to "public/people orientation focus". Part of the description of the technique "co-creation" was removed because it was repeated in the concept "relationship building. Some other texts were paraphrased. The adapted model is presented in Table 1.

Social Marketing Principle
• Aim: To bring social value through reciprocal exchange of resources or assets at the individual, community, societal or global level. • Social policy, strategy, products, services, and/or experiences are developed that will enable or assists publics to derive social benefits individually or collectively.

3-Social offerings
• Publics (citizens, policy-makers or stakeholders) are offered products, ideas, understanding, services, experiences, systems and environments that provide value and advantage. • Social offerings are in most cases positive in nature, although some can involve the imposition of restrictions that have collective support and benefit.

4-Relationship building
• Publics are engaged in the selection of priorities, and the development, design, implementation and evaluation of interventions.

1-Integrated intervention mix
• Strategies for change that include the marketing mix and other strategies. • Driven by target market insight data, segmentation analysis, competition analysis and feasibility analysis. • Mix of "types" and "forms" of interventions that are selected and coordinated to produce an effective and efficient program.

2-Competition analysis and action
• Internal (e.g. internal psychological factors, pleasure, desire, risk taking, genetics, and addiction, etc.) and external competition is assessed (e.g. economic, social, cultural and environmental influences). • Strategies are developed to reduce the impact of negative competition on the target behavior.

3-Systematic planning and evaluation
• Interventions use proven strategy and planning theory and models to construct robust intervention plans that include formative research, pretesting, situational analysis, monitoring, evaluation, and the implementation of learning strategies.

4-Insight-driven segmentation
• Segmentation using demographic, observational data and psycho-graphic data is used to identify groups that are similar and can be influenced in common ways. • Segmentation leads to the development of an intervention mix directly tailored to specific target market needs, values and circumstances.

5-Co-creation
• Strategies are developed to maximize the contribution of partner and stakeholder coalitions in achieving targeted behaviors.

Data items
Extracted information included: author(s), paper title, location, world region classified according to the six WHO regional groupings, country income category as classified by the World Bank, the aim of the intervention, NTD, population targeted, social marketing concepts, and social marketing techniques. Changes in behavioral determinants, behavior change, and health outcomes were also extracted. The behavior focus was categorized according to the five WHO public health strategic interventions for NTDs.

Quality assessment
The quality of included studies was assessed in the final stage of the review. The assessment was conducted using the QATSDD Critical Appraisal Tool [40] that assesses "the congruency, transparency and organized reporting of the research processes" [41] of studies of diverse designs (qualitative, quantitative, mixed). This tool was selected because it can be used to assess both quantitative and qualitative studies, and social marketing studies commonly use both.
The tool, used in other systematic reviews [42,43], comprises 16 criteria, each having a scoring scale from 0 to 3 (0 = not at all, 1 = very slightly, 2 = moderately, 3 = complete). Following the scoring guidelines [40] two reviewers, who did not participate in the screening or data extraction processes, assigned a score to the studies describing intervention results, calculated overall scores per intervention, and then consolidated results. To compare the quality across studies and following the QATSDD guidelines [40], the total scores of the interventions were converted to percentages. All studies were included independently of the appraisal results. This to have an overall understanding of the concepts and techniques incorporated by the interventions, and not limit findings only to studies of high-quality scores. Nevertheless, the results and implications of the quality assessment are presented in the following sections.

Synthesis of results
Extracted data were synthesized according to the five social marketing concepts, the four social marketing techniques, behavioral determinants, behavior changes, and health outcomes, and the five WHO public health strategic interventions for NTDs. Comparisons and meta-analysis were not planned nor conducted in this review.

Study selection
Out of 2,556 records screened, 57 full-text articles were retrieved to confirm eligibility. Additional publications (n = 23) complementing the interventions described in those 57 articles were retrieved through reference screening and Google searches. This led to the inclusion of 47 publications describing the 20 interventions that were included in the analysis. Most were published in English, with the exception of one in Spanish [44] (see Fig 1). As specified in the research protocol, a separate analysis of the gender responsiveness of the interventions included in this systematic review is reported elsewhere [45].
The results section is organized in four parts (see Fig 2). First, the characteristics of the interventions in terms of diseases of focus, the five WHO public health strategic interventions for NTDs they related to, and the countries of implementation are presented. Then, following the elements of the Hierarchical Model of Social Marketing, the concepts and techniques used by the interventions are highlighted. This is followed by a section focused on the outcomes reported by the interventions in terms of behavioral determinants, behavior change, and health outcomes. Finally, the success in reaching the outcomes is showcased. These last two
Five of the seven dengue interventions were conducted in Latin America, three of the four schistosomiasis interventions were implemented in China, and five of the seven interventions carried out in the Western Pacific region focused on WASH. See Table 2 for a list of the interventions.

Social marketing concepts and techniques used
The Hierarchical Model of Social Marketing concepts are: (1) social behavioral influence, (2) public / people orientation focus, (3) social offerings, and (4) relationship building. The five techniques listed in the model are: (1) integrated intervention mix, (2) competition analysis  Promote the use of the fivesteps cleaning method "La Untadita" to reduce mosquito infestation in cement washbasins and metal drums.
(Continued ) and action, (3) systematic planning and evaluation, (4) insight-driven segmentation, and (5) co-creation through social markets. See Table 3 for an overall summary and the supporting information (S2 File) for more details of the elements from the hierarchical model of social marketing included in the interventions. Concept 1: Social behavioral influence. This concept refers to behaviors or clusters of behaviors the interventions intended to influence, and the use of behavioral theories, analyses, and models to understand them [15]. All the interventions integrated characteristics of this concept (see Table 3, S2 File). The behaviors of focus of the interventions are consistent with four of the WHO public health strategic interventions for NTDs. Interventions addressing

PLOS NEGLECTED TROPICAL DISEASES
Social marketing to prevent and control neglected tropical diseases: A systematic review behaviors related to PCT focused mostly on influencing the target audiences to take treatment [47][48][49][50][51][52][53] as well as on general preventive measures [54], and increasing screening [53]. The intervention related to IDM focused on individuals with suspicious skin lesions to seek diagnosis and comply with multidrug therapy (MDT) for leprosy [51,52]. Prevention and control interventions related to VEM, focused on the control or elimination of mosquito breeding places [44,[55][56][57][58][59][60][61][62][63][64][65][66]. WASH behaviors focused on general preventive measures [47,54,, and more specific behaviors such as clean faces [69]. The interventions did not report their objectives following the SMART (specific, measurable, attainable, relevant, time-bound) structure, nor presented specific indicators. The behaviors of focus per intervention are listed in the supporting information (S2 File), as well as the stream levels the interventions focused on. Some publications describing the interventions explicitly mentioned the use of behavioral theories. One intervention focusing on schistosomiasis in China considered the Social Cognitive Theory of Multimedia Learning [71,72]. Another intervention addressing dengue in Colombia was designed following the stages of the Precaution Adoption Process Model [44]. Lastly, an intervention for Guinea-worm implemented in Nigeria, used the PRECEDE model [68]. Furthermore, the formative research conducted helped understand the contexts influencing people's behaviors. This is reported in the following concept.
Concept 2: Public / people orientation focus. This refers to understanding what influences people's behaviors (e.g. beliefs, attitudes, needs, wants) [15], through the collection and analysis of data using different research methods [26,29]. This informed understanding guides the development and implementation of actions along all phases and components of the intervention. This concept focuses on the research conducted to build that understanding of the audience, to ensure the interventions were oriented to the people.
Concept 4: Relationship building. This last concept is present along all stages of the intervention and entails engaging and having valuable exchanges with different stakeholders, including but not exclusively with the selected publics of the intervention [15]. Eighteen interventions provided evidence of processes of engagement of stakeholders in the selection of priorities, development, or implementation of the activities (see Table 3, S2 File). Some interventions initiated stakeholder engagement from early stages [53,55,57,71,72] and others benefited from previous experience in the settings [54,58,74]. Two interventions explicitly mentioned the engagement of intersectoral groups [53,57]. Government officials or health authorities were approached by nine interventions [44,[49][50][51][52][53]53,55,57,63,76]. Other publics were also engaged (see Table 4).
Technique 1: Integrated intervention mix. This refers to the types of strategies used to deliver value and influence behaviors [15]. These include the expanded marketing mix (i.e. product, price, place/distribution, promotion, partnerships, policy) and others such as advocacy, lobbying, and public and media relations [15,92]. All the interventions developed a mix of strategies that when integrated, provided value to the audiences (see Table 3, S2 File). The interventions included in this review used eight types of strategies: product, price, place, promotion, partnerships, policy, public relations, and community mobilization.
In social marketing, price refers to what the audience has to exchange or give up to receive the benefits offered by the intervention [28,93]. The element of price was considered by some interventions [49,52,58,[73][74][75][76]. This included the purchase of water filters [74], assuming the costs of materials and manpower [76,77], fear of adverse reactions to treatment [49], and psychological costs (e.g. fear of the disease) [51,52].
Formal partnerships were explicitly mentioned by three interventions [51,52,69,70,76,80,83]. The trachoma intervention targeting aboriginal communities had several, for example, one to develop culturally appropriate resources that involved, among others, past and present aboriginal health workers; and another one with the Melbourne Football Club (MFC) to allocate two Indigenous players as trachoma ambassadors [69,70,80,83].
Policy was mentioned by two interventions. One obtained political support from local authorities to facilitate garbage collection [55], and another was designed considering the provincial policy on health education for schistosomiasis [53].
Public relations activities were mentioned by two interventions [50,64,65]. One conducted meetings with State and District administrations that resulted in press releases that facilitated participant engagement [50], and the second one delivered invitations to each household inviting to participate in the final community meeting [64,65].

PLOS NEGLECTED TROPICAL DISEASES
Social marketing to prevent and control neglected tropical diseases: A systematic review addressed fear to adverse reactions to treatment [49], and another addressed misconceptions about leprosy, guilt, and shame by leprosy sufferers and their lack of acceptance of the biomedical explanations of the disease, and the difficulty to recruit health care staff due to stigma [51,52]. Other competing factors mentioned by the interventions included the existence of community misperceptions of effective mosquito breeding control measures [62,[64][65][66], preference to wells and alum instead of filters, and perception of water filters as having low efficacy [67,[73][74][75]; and expectations on government agencies solving problems and beliefs in traditional methods [68]. One intervention focusing on trachoma, challenged embarrassment and community shame about personal hygiene, and that 'dirty faces' were considered as normal in young children in remote indigenous communities by staff from clinics, schools, and community workplaces.
These organized and carried out clean-up campaigns and sewage maintenance work. the new toilet [76,77]. Dengue interventions evaluated different elements with regards to mosquitoes, including, vector densities [57,58], larval indices [55], presence of larvae [64], larval densities [63], and presence of intradomiciliary breeding places [44]. One intervention evaluated processes of empowerment, collaboration and mobilization [57] and another community mobilization [55]. Technique 4: Insight-driven segmentation. This technique refers to identifying in the study population subgroups (segments) of people with similarities, selecting among them segments to target, and developing specific strategies to influence their behaviors [15]. Seventeen of the 20 interventions applied this technique (see Table 3, S2 File). Evidence of the use of segmentation was provided for dividing the market into smaller segments [44,47,52,65,69,86], offering reasons for target audience selection [47,58,59,69,86], developing different strategies for varied groups of people [44,47,50,52,65,69,76,86], and by mentioning how insights from daily lives of the participants or from formative research were incorporated in the design or delivery of the intervention [44,48,49,53,55,58,65,72,73,89]. For example, one intervention mentioned that they used drama techniques (e.g. role-plays, storytelling) because they are part of African performance traditions [54]. A Guinea worm intervention promoting the purchase and use of monofilament nylon cloth filters allocated the salesforce to sectors considering the usual agenda of primary health workers (PHWs) [67,[73][74][75]. Another intervention classified participants according to the stages of the Precaution Adoption Process Model and implemented activities according to the stages at which they were [44].
Technique 5: Co-creation. This technique refers to strategies used to facilitate the participation of the main publics, for example, in the development of value propositions [15]. Twelve interventions incorporated this technique [27,[31][32][33]35,41,43,46,48,49,51,54] (see Tables 3 and  6, S2 File). Examples of co-creation strategies include having workshops with teachers and schoolchildren to learn how to develop educational videos and to create them, and feedback meetings at schools and communities to share results from formative research and to envision possible activities to undertake to reduce transmission [54]. Other examples are the use of community meetings to co-create messages and layout of pamphlets, and photo sessions to develop a photonovel using pictures from community members carrying out control activities [65].

Behavioral determinants, behavior changes, and health outcomes
Behavioral determinants are considered mediating outcomes to influence behavior [17]. Changes in behavior are needed to attain positive health outcomes (e.g. reducing infection rates). In this review, the five public health strategic interventions prioritized by the WHO are used to present the behavior focus and outcomes reached by the interventions. These outcomes were not predefined by the study; instead, these were coded inductively according to what the interventions reported. Due to the variety of outcomes considered in this review, comparisons nor meta-analysis were carried out.

Self-reported results of the interventions
Behavior influence is central to social marketing [15,26,27]. According to the evaluation results reported by 19 interventions, overall, they were successful in changing behaviors related to

PLOS NEGLECTED TROPICAL DISEASES
Social marketing to prevent and control neglected tropical diseases: A systematic review NTD prevention and control. A direct relationship between the number of concepts or techniques used and the success of the interventions in reaching the planned outcomes was not observed. It was not possible to quantify the effectiveness of the interventions due to the heterogeneity of the measures and outcomes of the interventions. For this reason, it was also not feasible to present effectiveness according to the concepts and techniques used. However, the interventions were successful in achieving positive outcomes, and these are reported in this section following the five WHO public health strategic interventions for NTDs (see S3 File). The interventions focusing on PCT were able to increase the number of participants receiving [50], and taking treatment [48][49][50]53], MDA coverage [48], and adherence to screening [53]. These interventions used at least three concepts and two techniques.
The intervention focusing on IDM increased the number of individuals seeking treatment and early detection [52]. It reported a decrease in the national prevalence rate for leprosy and was successful in reaching most of its objectives. It used all concepts and techniques, except for the technique co-creation.
Interventions focusing on VEM were successful in eliminating breeding places [44,64] and reducing vector densities [55,57]. One reported a relative reduction of larval infestation indices in the intervention neighborhoods [63]. The "La Untadita" intervention reported that changes in human behavior were achieved, but the level of reduction in the overall infestation indices was not enough for adequate control of mosquitoes [58]. The interventions not conducting entomological evaluations reported increasing the number of participants in the stages six (action) and seven (maintenance) of the Precaution Adoption Process Model [44], and a positive impact on knowledge and practices related to dengue fever (DF) prevention [59]. The interventions focusing on VEM [44,55,[57][58][59]63,64] used at least two concepts and two techniques.
Interventions focusing on WASH were able to motivate the communities to build toilets and use them [77] and to purchase and use water filters [75]. These interventions successfully prompted children to clean their faces [69,86], hand wash after toilet use [71], and decrease contact with unsafe water [89]. In one intervention, villagers adopted preventive measures, and village health volunteers referred cases to the health center [68]. An intervention addressing soil-transmitted helminths "was associated with 50% efficacy (95% CI, 30 to 65) in preventing infection" [71]. Of these, almost all used the four social marketing concepts and four [69][70][71][72]76,86] or five [68,[73][74][75] techniques.
Of the two interventions using combined PCT and WASH strategies, one only reported behavioral determinants [54]. The other one was successful in decreasing the frequency of water contact among women and children, but not among men, and increased compliance with chemotherapy in the three segments [47]. Both applied four concepts and three techniques, none addressed competition.

Methodological quality
According to the results of the QATSDD critical appraisal tool [40], the average quality score for all papers describing the interventions was 62.5% (see Fig 3). The methodological quality scores ranged from 20.8% [51,52] to 97.9% [71,72]. Most of the interventions (n = 12) achieved a score of 60% or higher (see Fig 3, S4 File). The lowest scores were reported on the criteria "assessment of reliability of analytical process" not applicable for quantitative studies, and "statistical assessment of reliability and validity of measurement tool(s)" not applicable to qualitative studies. Criteria for which most studies score high were "statement of aims/objectives in main body of report" and "clear description of research setting", both applicable to quantitative, qualitative, and mixed methods studies. Two interventions achieved scores lower than 50%, namely the intervention addressing leprosy in Sri Lanka [52], and an intervention focused on schistosomiasis implemented in China [89]. The former reported positive changes in behavioral determinants, behavior change, and health outcomes (i.e. prevalence), but did not report their methodological decisions and processes thoroughly. The latter received low scores in most of the criteria assessed, mainly for providing minimal descriptions and justification of their methodological processes. In contrast, only one intervention achieved a quality score higher than 80% [71,72]. This intervention focused on soil-transmitted helminths in China, obtained a 97.9% score. It thoroughly reported the methods used and rationale, as well as provided data on behavioral determinants, behavior change, and health (i.e. incidence) outcomes.

Discussion
Social marketing interventions for the prevention and control of NTDs have focused on eight of the 17 diseases prioritized in the WHO NTD Roadmap [34]. Although soil-transmitted helminths are the largest cause of NTD-related disability-adjusted life-years (DALYs), only one of the 20 interventions focused on this disease, whereas seven focused on dengue which is only the sixth leading cause of NTD-related DALYs [94]. The 20 interventions relate to four out of five of the WHO priority strategies for preventing and controlling NTDs, with one intervention addressing behaviors related to IDM, and none focusing on veterinary public health measures. Although NTDs represent a major burden in sub-Saharan Africa [94,95], only three of the 20 interventions identified were from this region. To our knowledge, this review is the first to use the Hierarchical Model of Social Marketing [15]. Overall the interventions integrated most concepts and techniques of the model. More than 15 interventions incorporated the concepts "relationship building" and "public / people orientation focus", and the technique "insight-driven segmentation". Only the technique "competition analysis and action" had less than ten interventions applying it.
A direct comparison with findings from other reviews is not possible, except to the extent that some elements of the Hierarchical Model of Social Marketing are similar or equivalent to the social marketing benchmarks [26,29,30] on which the model is grounded. The concept "public / people orientation focus" is equivalent to the benchmark "citizen orientation". Other social marketing systematic reviews of health interventions also found that the interventions well incorporated this benchmark [17,24]. The technique "insight-driven segmentation" is equivalent to the benchmarks "insight" and "segmentation". These benchmarks have also been among the most commonly used in social marketing global health interventions [17]. The technique "competition analysis and action" is equivalent to the benchmark "competition". Other systematic reviews of social marketing also found a low level of adherence to this benchmark [17,24,35,36]. There is no equivalent benchmark to the concept "relationship building".
Social marketing concepts and techniques should be adapted to the specific characteristics of the context intervened. Nevertheless, some general insights that could be considered by any social marketing NTD intervention are drawn from the studies included in this review. One is that all actions and their outputs should be guided by a strong understanding of the audience and the local context. This was an important characteristic of the interventions included in this review. Other systematic reviews of social marketing have also highlighted the benefits of understanding the publics before developing interventions, such as being better equipped to influence behaviors [17,19]. Therefore, NTD social marketing interventions should understand the audience and context before designing possible solutions.
Another social marketing concept to highlight from the findings of this study is relationship building. Engaging audiences from the early stages of the intervention could be favorable to achieve NTD prevention and control outcomes. Primary beneficiaries, as well as other stakeholders, could cooperate in designing and delivering parts of the intervention [96]. This was observed in several interventions included in this study. Their involvement could help develop a sense of ownership [8,97], design realistic solutions based on local knowledge and capacities [98], and facilitate the sustainability of actions [55,97]. This is consistent with recommendations from other studies and reports calling for placing engagement of communities at the center [98], as well as for promoting intersectoral and cross-sectoral collaboration to prevent, control, and eliminate NTDs [8,99].
Regarding social marketing techniques, the intervention mix of strategies should be integrated. The strategies should not be implemented in isolation, but rather as complementary to each other. The interventions assessed in this review combined varied strategies to achieve their planned outcomes. This is something to replicate by other social marketing interventions. Additionally, taking into consideration internal and external factors that impede audiences adopting the desired behavior is important. Considering competing factors such as embarrassment, shame, and misconceptions helped some interventions in this review attain their planned outcomes. However, analysis of the competition and acting upon that was a missed opportunity for most interventions reviewed. Additionally, NTD social marketing interventions that showed positive results, more often than not, integrated health education, had a capacity building component and were culturally appropriate. These elements should also be considered in planning and delivering future interventions.
This study identified the need to expand the application of social marketing to address behaviors related to IDM and veterinary public health measures. Most of the interventions included in this review focused on behaviors associated with the priority public health strategic interventions that receive more considerable attention from the NTD community and funders, namely PCT, VEM, and WASH. The focus on these could be the result of imbalances observed at the global policy level, where diseases that were once considered tool-deficient were found to be less prioritized [100]. Therefore, moving IDM and veterinary public health measures up in the global policy agenda could help in driving more resources to implement social marketing interventions focused on behaviors related to these strategic interventions.
Social marketing interventions can focus simultaneously on more than one of the five priority strategies to prevent and control NTDs. Although this study only found two interventions combining PCT and WASH strategies to address schistosomiasis, other interventions could have also benefited from using multiple priority strategies. Only one intervention addressing schistosomiasis screened and provided treatment for intestinal helminths in parallel to the disease of focus, but it did not implement measures beyond that, missing the opportunity to address both diseases simultaneously. The benefits of clustering NTDs have been highlighted in the literature, including its cost-effectiveness [103]. Combining strategies may provide more effective control of one or several diseases, even if one strategy dominates [101,102].
Only four of the 20 interventions self-identified as social marketing interventions. This raises questions as to whether public health programs use social marketing without knowing it or decide not to label it as such. The challenges in identifying social marketing interventions have been raised by other scholars [17,19,104,105]. Quinn et al. [105] found that some selflabeled social marketing interventions were only focused on disseminating health messages and, therefore, were not social marketing. While conducting reviews, other scholars have found that the characteristics of the interventions rather than the label help determine whether they have applied social marketing elements [104]. An establishing standard and measurable criteria to operationalize the social marketing definition and elements (e.g. benchmark, concepts, techniques) is warranted.
According to the QATSDD critical appraisal tool, overall, the studies in this review were of variable quality. For this reason, the outcomes reported should be considered with caution, especially for studies with lower quality scores. Social marketing interventions addressing NTDs need to improve their quality of report and provide more details about the methodological rationale guiding their actions, as specified in the QATSDD.
Behavior change is necessary to attain NTD outcomes. This has been highlighted in reports of progress towards NTD control, elimination and eradication targets [8,101,102,106]. The five WHO public health strategic interventions involve influencing behaviors of downstream (individual), midstream (e.g. communities), and upstream (e.g. policy makers) audiences. Social marketing as an approach to behavior change can contribute in promoting voluntary behavior change at all these levels [16,107]. Incorporating social marketing elements in developing interventions for NTDs would also be aligned with the new NTD roadmap for the period 2021-2030, prioritizing integrated approaches and cross-sectoral collaboration [108]. This has also been postulated at the global NTD policy level, more precisely by the Director General of Mexico's National Center for Disease Prevention and Control during the 2017 Global Partners Meeting on NTDs when he said: "for us, it is important to have a tactical approach and continuous action in epidemiological surveillance, entomological surveillance, health promotion, and social marketing, insecticide resistance and monitoring that resistance, the vector control, and the health care guidelines" [109].
However, NTD policies do not explicitly mention social marketing as an approach to behavior change as do other national and global health policies or reports suggesting its use for health promotion and disease prevention [110][111][112][113][114][115]. While further research is warranted, the interventions reviewed in this study show that social marketing has been used successfully in NTD prevention and control. Our findings show that social marketing is aligned with the current priority strategies for the prevention and control of NTDs. Thus, ongoing discussions to formulate the targets and milestones of the new global Roadmap for NTDs, presently under development by the WHO in consultation with the NTD community, should integrate social marketing as an approach to overcome these diseases.

Strengths and limitations
The use of the Hierarchical Model of Social Marketing [15] for data extraction and analysis was both a strength and a limitation in this study. It is the first time it was used in a systematic review of social marketing, and while this is thus innovative, this innovation meant a lack of precedent, and we found overlapping concepts and techniques in the model. An additional limitation was found in the difficulty to operationalize the elements of the framework due to their broad definitions. The challenges in operationalizing social marketing criteria have been raised in other studies [17,19,104]. It is recommended that future systematic reviews using the Hierarchical Model of Social Marketing, test and develop tools and methods to operationalize the concepts and techniques better. In doing this, studies should adapt the model to respond to the context of LMICs, settings with the highest burden of NTDs.
Moreover, this review focused on the 17 NTDs prioritized by the WHO at the time of study design, three new diseases (i.e. chromoblastomycosis and other deep mycoses, scabies and other ectoparasites, snakebite envenoming) added to the WHO NTD portfolio as of 2017 were not included. The reason for this was that these diseases were included after the study selection process was finalized, and when data extraction was in progress.
A notable limitation in this study was the absence of a validated measurement tool for the Hierarchical Model of Social Marketing. Similarly, this review is limited by the fact that social marketing interventions may have been missed simply because they lacked explicit terminology related to social marketing. We aimed to compensate for these by having experts in social marketing involved in the study selection and coding processes, and by using broad terms in the search strategies. These strategies did not restrict papers to any specific language. Where data were missing, we contacted the authors of the papers. However, not all authors we contacted in the study selection process responded to inquiries, and due to missing information, some interventions could not be included. Therefore, as reported in the PRISMA diagram, this review did not include some interventions due to missing data.

Conclusion
NTDs have gained recognition globally as a result of the interplay of various determinants of political priority [100]. These include having a common set of goals and the involvement of actors from multiple sectors to reach them [8,34,100]. Progress has been made to reach global NTD targets, but collaboration across sectors and disciplines has been highlighted as fundamental to attain global goals [8,116]. Social marketing can contribute in reaching NTD targets. Evidence from this study shows that social marketing has been successfully used to address behaviors related to most of the public health strategic interventions for NTDs.
Moving forward, social marketing interventions for the prevention and control of NTDs need to be grounded on an understanding of the audience and adapted to the contexts. They should also build stakeholder relationships as early as possible and involve the publics. Elements of the intervention mix should be integrated and mutually supportive. Incorporating health education and capacity building, as well as being culturally appropriate, is also important. Challenges that remain include expanding the application of social marketing to address behaviors related to IDM and veterinary health measures as well as addressing several diseases with similar characteristics simultaneously, and combining NTD priority strategies. Addressing these challenges, building from positive experiences highlighted in this review, could be beneficial to attain NTD prevention and control outcomes.