## Abstract

Social marketing is a commonly used strategy in global health. Social marketing programmes may sell subsidized products through commercial sector outlets, distribute appropriately priced products, deliver health services through social franchises and promote behaviours not dependent upon a product or service. We aimed to review evidence of the effectiveness of social marketing in low- and middle-income countries, focusing on major areas of investment in global health: HIV, reproductive health, child survival, malaria and tuberculosis. We searched PubMed, PsycInfo and ProQuest, using search terms linking social marketing and health outcomes for studies published from 1995 to 2013. Eligible studies used experimental or quasi-experimental designs to measure outcomes of behavioural factors, health behaviours and/or health outcomes in each health area. Studies were analysed by effect estimates and for application of social marketing benchmark criteria. After reviewing 18 974 records, 125 studies met inclusion criteria. Across health areas, 81 studies reported on changes in behavioural factors, 97 studies reported on changes in behaviour and 42 studies reported on health outcomes. The greatest number of studies focused on HIV outcomes (n = 45) and took place in sub-Saharan Africa (n = 67). Most studies used quasi-experimental designs and reported mixed results. Child survival had proportionately the greatest number of studies using experimental designs, reporting health outcomes, and reporting positive, statistically significant results. Most programmes used a range of methods to promote behaviour change. Programmes with positive, statistically significant findings were more likely to apply audience insights and cost-benefit analyses to motivate behaviour change. Key evidence gaps were found in voluntary medical male circumcision and childhood pneumonia. Social marketing can influence health behaviours and health outcomes in global health; however evaluations assessing health outcomes remain comparatively limited. Global health investments are needed to (i) fill evidence gaps, (ii) strengthen evaluation rigour and (iii) expand effective social marketing approaches.

Key Messages

• Social marketing is commonly used as an intervention strategy in global health, but there have been few systematic reviews of its effectiveness in changing behaviours or influencing health outcomes in low- or middle-income countries.

• We systematically reviewed peer-reviewed literature published between 1995 and 2013 on social marketing for HIV, reproductive health, malaria, child survival and tuberculosis in low- and middle-income countries to assess the capacity of social marketing to achieve improvements in health outcomes, health behaviour change and behavioural factors

• Of 125 studies meeting our inclusion criteria, one-third (n = 42) reported on measureable health outcomes, while most measured behavioural outcomes (n = 97) or behavioural factors (n = 81). Key strategies of effective programmes included using research to seek actionable insights into consumers and markets and addressing the costs and benefits of behaviour change.

• Evidence exists that social marketing can influence health behaviours and health outcomes. Better documentation of social marketing interventions is needed when evaluating these programs. Global health investments are merited in filling key evidence gaps, such as in social marketing to address childhood pneumonia, and to strengthen evaluation rigour. Investments are needed as well in scaling up and replicating effective social marketing approaches.

## Introduction

Social marketing is commonly used as an intervention strategy in global health. Social marketing uses marketing concepts—product design, appropriate pricing, sales and distribution, and communications—to influence behaviours that benefit individuals and communities (Cheng et al. 2011). Social marketing interventions generally aim to ensure that the target audience adopts the behaviour being promoted (Lee and Kotler 2011). Social marketing supports markets to make health products and services appealing and affordable for both health care providers and consumers (Meadley et al. 2003; Cairns et al. 2011; Berg and Mitchell 2013). Social marketing programmes sell behaviours. This often includes products that support behaviours by subsidizing them and making them available at commercials outlets, but the purpose of making a product or service available is to support the practice of the behaviour being targeted (Cheng et al. 2011). In low- and middle-income countries, these programmes may sell subsidized products through commercial sector outlets such as pharmacies, distribute appropriately priced products to the poorest populations, deliver health services through social franchises or promote behaviours not dependent upon a product or service. Social marketing may also operate in a more upstream manner to address policy or social norms change (Andreasen. 2002; Gordon et al. 2006; Weinreich 2006; Stead et al. 2007). Social marketing is commonly used as an approach in global health to reach vulnerable populations with health products and services (Price 2001; Mah et al. 2008; Berg and Mitchell 2013).

Although social marketing has been included in investments made within the context of development assistance for health, the effectiveness of social marketing for achieving improvements in major global health indicators has been questioned (Andreasen 2002; World Health Organization 2014). Promoters and detractors of social marketing may both agree on the ultimate goal of achieving behaviour change in target populations; however critics of social marketing have argued that behaviour change is difficult to achieve solely through the mechanisms adopted by social marketing (Buchanan et al. 1994; Lee and Kotler 2011). Even if social marketing may encourage people to try marketed products and services, it is not clear that this behaviour change can be sustained. Definitions of social marketing have expanded in response to critiques that social marketing can be effective only if the approach operates within an enabling policy and social environment (Buchanan et al. 1994; Andreasen 2002; Hornik 2002; Smith and Schneider 2009; Spotswood et al. 2012). Social marketing definitions now include more explicit mention of the linkage to policy and advocacy (Stead et al. 2007). Others critique social marketing from a view of informed consent, asking whether having a socially beneficial end goal always justifies nudging audiences, especially vulnerable populations, toward behaviours they otherwise would not choose to adopt (Guttman and Salmon 2004; Spotswood et al. 2012).

A key issue in assessing the effectiveness of social marketing is whether social marketing programmes actually achieve behaviour change on a consistent basis. Programmes employing mass media or interpersonal communication channels may raise the target audience’s awareness and intentions to use products and services. These behavioural factors—improved attitudes, knowledge or perceived self-efficacy—may facilitate the end goal of social marketing, but do not provide sufficient evidence that social marketing interventions have achieved their stated purpose. Rather, these factors may mediate the process of behaviour change or can be categorized as intermediate outcomes (Andreasen 2002). Demonstrated evidence of changes in behaviour or improvements in health status could address these concerns and provide a more reliable assessment of effectiveness and impact (Raphael 2000).

### Existing systematic reviews of social marketing

We identified six systematic reviews that touch on the effectiveness of social marketing for global health investments in low- and middle-income countries (Bertrand et al. 2006; Bhutta et al. 2008; Sweat et al. 2012; Beyeler et al. 2013; Evans et al. 2014; Naugle and Hornik 2014). A common feature of these and other reviews of social marketing, including of studies taking place in high-income countries, is the challenge of identifying social marketing interventions and the risk of mis-classifying social marketing as health promotion or social and behaviour change communication (McDermott et al. 2005; Stead et al. 2007,; Quinn et al. 2010).The six reviews we found focusing on low- and middle-income countries used varied definitions of social marketing. Several considered social marketing as embedded within mass media interventions and defined the aims of these interventions as being to raise awareness or change behaviours among a broad audience (Bertrand et al. 2006; Bhutta et al. 2008; Noar et al. 2009; Quinn et al. 2010; Naugle and Hornik 2014). Others argued that social marketing comprises a broader set of intervention components, acknowledging that social marketing may operate through a marketing mix, i.e. factoring in pricing and sales strategies (Sweat et al. 2012; Evans et al. 2014). Many practitioners acknowledged that it difficult to determine whether interventions that identify as ‘social marketing’ are actually rooted in principle (Quinn et al. 2010). As a result of the varied definitions of social marketing, it is possible that these reviews inadvertently included behaviour change and health communications campaigns alongside more strictly defined social marketing programmes. Loose applications of the principles of social marketing as an intervention strategy make it difficult to trace the theoretical pathways by which social marketing achieves behaviour change and improved health outcomes, particularly for those health outcomes of greatest interest to the global health policy community.

These reviews found evidence of positive effects of social marketing campaigns on the target population’s behaviours, intentions to adopt healthy behaviours and factors mediating the process of behaviour change, such as awareness, knowledge and attitudes. However, studies tended to report more often on these mediating factors, or intermediate outcomes, than on behaviours. Studies that measured both types of outcomes often found that exposure to the intervention produced stronger associations with improvements in these mediating factors than in behaviours (Price 2001,; Bertrand et al. 2006; Evans et al. 2014). Of three HIV-focused reviews, two were able to draw conclusions about the effects of social marketing on changing behaviour, finding that social marketing was associated with increases in condom use (Noar et al. 2009; Sweat et al. 2012). A review of social franchising, a sub-set of social marketing, found that franchised reproductive health clinics were linked to improvements in service utilization and uptake of family planning among clients (Beyeler et al. 2013). A recent review of social marketing of water and sanitation found mixed evidence of behaviour change related to water treatment (Evans et al. 2014). Two reviews that focused on social marketing as one of a range of delivery strategies assessing the effectiveness of behaviour change interventions related to reproductive or maternal and child health care outcomes did not present data specifically on the ability of social marketing to achieve behavioural outcomes (Peters et al. 2004; Bhutta et al. 2008). Authors have generally characterized the evidence reviewed as being inconsistent and of weak quality.

None of these reviews have addressed the full range of products and services for which social marketing may be mobilized in low- and middle-income countries. As a result, it is difficult from these reviews to assess the ability of social marketing as a strategy to contribute to meeting the health Millennium Development Goals (MDGs), which have set the course for global health investments, particularly development assistance, since 2000, and are still relevant as the global health field moves into targeting the post-2015 Sustainable Development Goals (UN Millennium Project 2005; Institute for Health Metrics and Evaluation 2014). Critically, none of these reviews has assessed whether social marketing programmes can achieve improvements in health status while also addressing whether social marketing can consistently change behaviour.

We aimed to review available evidence on the effectiveness of social marketing to achieve improvements in health outcomes, health behaviour change and mediating behavioural factors in low- and middle-income countries, focusing on major areas of investment in global health by multilateral and bilateral donors, as evidenced by the presence by multilateral investment frameworks and large-scale investment by bilateral agencies and/or private foundations focused on achieving the health MDGs (Partnership for Maternal, Newborn, & Child Health 2011; The Global Fund to Fight AIDS, Tuberculosis, and Malaria 2011; Joint United Nations Programme on HIV/AIDS (UNAIDS) 2012; Family Planning 2020 2014; Stenberg et al. 2014). We used a logic model to conceptualize a generalized pathway for the expected effectiveness of social marketing (Figure 1). This model posits that a social marketing program, which may include many components of the marketing mix—product, price, promotion, place—and may involve both supply- and demand-side strategies—should lead to individual-level programme exposure by the target audience. Individual-level exposure should influence a change in mediating behavioural factors—access and availability to health products and services, knowledge, attitudes, social norms, intentions etc.—that will ultimately lead to adoption of a health-promoting behaviour. Healthy behaviour change (i.e. taking preventative and/or treatment action), should then lead to improvements in health status, as assessed through measures of morbidity, mortality, or fertility status (Glanz et al. 2002; Lee and Kotler 2011).

Figure 1.

Logic model on the effects of social marketing.

Figure 1.

Logic model on the effects of social marketing.

## Methods

### Eligibility criteria

This review was structured in accordance with the PRISMA statement (Liberati et al. 2009). Social marketing programmes were defined as programmes using marketing concepts to develop activities aimed at influencing people’s behaviours to improve the public’s health, assessed through application of the social marketing benchmark criteria developed by the National Social Marketing Centre (Cheng et al. 2011; The National Social Marketing Centre 2012). Studies were included if they: (i) were original research (not editorials or review papers that did not contain original data); (ii) took place in a low- or middle-income country as defined by the World Bank; (iii) published in English; (iv) assessed an programme that attempted to change a behavioural factor, behaviour or health outcome of interest within the global health investment areas of HIV, reproductive health, child survival, malaria and tuberculosis; (v) provided adequate information to determine if the programme met at least one of the Social Marketing Benchmark Criteria; (vi) and used an experimental, including randomized-controlled trials or quasi-experimental study design (World Bank Group 2016).

Because previous reviews had identified few experimental studies of social marketing, and evaluation designs varied greatly in rigour, we aimed to be inclusive in our identification of eligible study designs. We defined studies as quasi-experimental if they had either a comparison group or both pre- and post-tests, and we excluded one-group post-test only designs, as they can only in rare circumstances demonstrate whether a change has occurred (Shadish et al. 2002). Economic evaluations were excluded, unless it was possible to identify behavioural factor, behaviour or health outcome results.

Health outcomes included measures of morbidity, mortality or fertility status. Specifically by health area, HIV health outcomes of interest included incidence and prevalence of HIV and sexually transmitted infections (STIs), as well as cure rates of STIs. Health outcomes of interest for reproductive health included reduction in fertility, maternal mortality and infant mortality. Malaria health outcomes included reduction in malaria incidence and prevalence, and malaria-related mortality. TB health outcomes included TB case identification and cure rate and TB-related mortality. Child survival health outcomes included incidence, prevalence and mortality due to diarrhoeal disease and pneumonia, undernutrition and stunting, anaemia, low birthweight, postnatal sepsis, maternal mortality and neonatal and under-5 child mortality.

Behavioural outcomes were defined as behaviours that could influence the aforementioned health outcomes. HIV/STIs behaviour outcomes included condom use, delayed sexual onset, HIV testing, partner reduction, male circumcision, safe injecting behaviours and HIV/STI treatment. For reproductive health, we categorized behaviour outcomes into client and provider outcomes. Client behaviour outcomes included modern contraceptive use, birth spacing, safe abortion, antenatal care, postnatal care and institutional delivery. Provider behaviour outcomes included quality of care (information given to clients, technical competence, interpersonal relations, follow-up/continuity mechanisms, methods mix) and client satisfaction (Bruce 1990). TB behavioural outcomes included TB testing, initiation and completion of Directly Observed Treatment Short course. Malaria behavioural outcomes included sleeping under an insecticide-treated net (ITN) or long-lasting insecticidal nets (LLIN), use of insecticide retreatment tablets or sprays, seeking care for fever and initiation and completion of antimalarial treatment, specifically artemisinin-combination therapies (ACTs). Malaria provider behaviour change outcomes included parasitological diagnosis using rapid diagnostic tests (RDTs) and provision of ACTs. Child survival behavioural outcomes included hand washing and other hygiene practices for prevention of both pneumonia and diarrhoea, pneumococcal vaccination and zinc supplementation and antibiotic treatment of pneumonia, water purification, use of latrines, and rotavirus vaccination. Other behaviours included use of oral rehydration solution (ORS) and/or zinc, antibiotic treatment for bacterial dysentery, complementary feeding including micronutrient supplementation and fortification, consumption of micronutrient-rich foods, therapeutic feeding including Ready-to-Use Therapeutic Foods and formula milks, and exclusive breastfeeding from 0 to 6 months. Behaviours to reduce neonatal mortality included use of iron folic acid and supplementation, use of clean delivery kits or chlorhexidine, misoprostol, oxytocin, magnesium sulfate and having a skilled attendant at birth.

Behavioural factors were defined as those conditions that affect the likelihood that an individual would perform a behavioural outcome of interest. Behavioural factors for all health areas included opportunity, ability and motivation to change health behaviours. Opportunity was defined as availability, access and brand equity; ability as knowledge, self-efficacy, social support and stigma; and motivation as attitudes, intention, risk perception and quality of care (Population Services International 2004).

### Search strategy

We searched for peer-reviewed studies published from 1995 to 2013 that evaluated the effectiveness of social marketing programmes in the areas of HIV, reproductive health, child survival, malaria and tuberculosis using the bibliographic databases of PubMed, PsycInfo (via Ovid) and ProQuest. Search terminology is described as follows:

‘Social franchise’ OR ‘Social franchises’ OR ‘Social franchising’ OR ‘Social Marketing Theory’ OR [‘Social Marketing’ AND (Health OR ‘HIV’ OR ‘AIDS’ OR ‘STI’ OR ‘sexually transmitted infections’ OR ‘TB’ OR Tuberculosis OR ‘Reproductive Health’ OR ‘Family Planning’ OR Contraception OR Condom OR ‘IUD’ OR ‘Maternal Health’ OR ‘Women’s Health’ OR Youth OR ‘Adolescent Health’ OR Malaria OR ‘Integrated Case Management’ OR fever OR artemisinin OR ‘artemisinin-based combination therapy’ OR ‘artemisinin-based monotherapy’ OR ‘ACT’ OR antimalarial OR ‘insecticidal bed nets’ OR bednets OR nets OR ‘ITN’ OR ‘LLIN’ OR ‘insecticide retreatment tablets’ OR ‘Rapid diagnostic test’ OR ‘RDT’ OR ‘Child Survival’ OR ‘Children’s Health’ OR ‘Infant Health’ OR ‘Oral Rehydration’ OR ‘ORS’ OR ‘ORT’ OR Salts OR Zinc OR ‘Water Treatment’ OR ‘Water Purification’ OR ‘Water Filters’ OR ‘Water Filtration’ OR Chlorination OR Chlorine OR Latrines OR Sanitation OR Soap OR Handwashing OR ‘Hand washing’ OR Nutrition OR supplementation OR ‘micronutrient powder’ OR fortification OR sprinkles OR ‘RUTF’ OR ‘RUSF’ OR ‘Formula milks’ OR ‘iodized salts’ OR vitamin OR ‘complementary feeding’ OR ‘therapeutic feeding’ OR ‘LNS’ OR ‘lipid-based nutrient supplements’ OR Iron OR ‘folic acid’ OR ‘iron folic acid’ OR prenatal OR antenatal or ‘low birth weight’ OR ‘Neglected Tropical Disease’ OR ‘Soil Transmitted Helminthes’ OR Deworming OR Pneumonia OR ‘Pneumonia treatment’ OR ‘Delivery Kits’ OR Chlorhexidine OR CHX OR sepsis OR vaccine OR vaccination)]

After duplicates were removed, two authors screened abstracts according to eligibility criteria. Full texts of studies were reviewed by three authors, and discrepancies were addressed through discussion.

### Data abstraction

Key variables were extracted from eligible studies, including: study location, study population, intervention description, study outcomes, design, sampling methods and statistical analysis applied. To assess the core social marketing components of included interventions, eligible studies were scored against the following criteria defined by the National Social Marketing Centre (The National Social Marketing Centre 2012): (i) Behaviour: does the intervention aim to change people’s actual behaviour vs their knowledge, attitudes, and beliefs? (ii) Customer Orientation: does the intervention focus on the audience and use a mix of data sources to fully understand their lives, behaviour, and the issue? (iii) Theory: does the intervention use behavioural theories to understand behaviour and inform the intervention? (iv) Insight: did customer research identify actionable insights that were used to develop the intervention? (v) Exchange: does the intervention consider the benefits and costs of adopting and maintaining a new behaviour? (vi) Competition: does the intervention seek to understand what competes for the audience’s time, attention, and inclination to behave in a particular way? (vii) Segmentation: does the intervention identify audience segments, which have common characteristics, and then tailor interventions appropriately? (viii) Methods Mix: does the intervention use a mix of methods to bring about behaviour change?

We scored studies against the eight benchmark criteria using a process similar to that of Quinn and colleagues in their 2010 review of the design of social marketing interventions (Quinn et al. 2010). Each study was awarded an equally weighted point for each of the benchmark criteria identified in the description of the intervention, for a maximum score of eight. Our intent in structuring this scoring process was to assess how well studies that were identified as social marketing interventions based on our search terms and eligibility criteria compared against the key characteristics of a well-designed social marketing program. Earlier reviews of social marketing, several of which included social marketing programmes from high-income countries, found that only a small proportion of their included studies explicitly mentioned all eight criteria (Gordon et al. 2006; Stead et al. 2007; Quinn et al. 2010). To account for wide differences in how interventions that identified as social marketing were described, we retained all included studies for analysis. We then developed a sub-analysis of studies that were scored as having six or more of the benchmark criteria and thus could be considered to have more rigorously implemented and/or reported on their use of social marketing principles (Quinn et al. 2010).

### Assessment of study rigour

A strength of evidence grade was applied to individual studies to assess and compare validity of study findings. The 5-point grading scale was adapted from the strength of evidence framework in the Cochrane Handbook (Higgins and Green 2015). Each individual study was graded based on the measured outcomes, study design, generalizability and type of analysis used.

### Analysis

Given the range of outcomes considered, we did not attempt meta-analysis. Rather, this review was designed to report on the breadth of the literature and indication of social marketing’s effectiveness based on the direction and significance of results. We categorized studies by the direction and statistical significance of reported effect sizes, stratifying results by outcome type, study design, and health area. Study effects were categorized as ‘positive’ if all results reported were statistically significant and improved outcomes. We categorized studies as ‘mixed’ if results had any combination of the following (excepting positive and statistically significant): positive, negative, statistically significant, and not statistically significant. For example, many included studies had a mix of statistically significant, positive results and results that were positive but not statistically significant. Within the mixed category, we also included studies that reported statistical significance for some results and statistical non-significance for others. The remaining studies were categorized into ‘not statistically significant/significance not reported’ if all results were either not statistically significant or if authors did not report significance for any of the results. No studies reported only statistically significant negative results across all three outcome types; however statistically significant negative results are reported by individual outcome types. We report counts of all categorized studies.

We also report proportions of Social Marketing Benchmark Criteria for all included studies, in total and stratified by outcome type. Finally, we assessed evidence gaps within the health areas reviewed. For each health area we identified evidence gaps for the types of outcomes reviewed, according to whether there were fewer than two studies reviewed in that category (Ryan 2013).

## Results

A total of 18 947 records were located using the described search terms. Of those, 1,443 duplicates were removed. Next, we reviewed 17 504 records at the abstract level. 16 936 studies were then eliminated based on the above inclusion criteria. Full-text review further removed another 440 studies, yielding 125 eligible studies for inclusion in analysis (Figure 2, see also

) (Alisjahbana et al. 1995; Schopper et al. 1995; Pinfold and Horan 1996; Agha 1998; 2002, 2003, 2011; De Pee et al. 1998; Valente and Saba 1998; Ford and Koetsawang 1999; Pinfold 1999; Schellenberg et al. 1999, 2001; Laukamm-Josten et al. 2000; Meekers 2000; Van Rossem and Meekers 2000, 2007; Vaughn et al. 2000; Abdulla et al. 2001; Agha et al. 2001, 2006, 2007a,b; Babalola et al. 2001a, b; Curtis et al. 2001; Dunston et al. 2001; Karlyn 2001; Jaramillo 2001; Kim et al. 2001, 2006; Boulay et al. 2002; Rowland et al. 2002, 2004; Thevos et al. 2002; Collumbien and Douthwaite 2003; Jacobs et al. 2003; Tambashe et al. 2003; Zagré et al. 2003; Basu et al. 2004; Nathan et al. 2004; Shefner-Rogers and Sood 2004; Warnick et al. 2004; Angeles-Agdeppa 2005; Crape et al. 2005; Goldstein et al. 2005; Kanal et al. 2005; Khan et al. 2005; Kikumbih et al. 2005; Mathanga et al. 2005; Meekers et al. 2005; Paulino et al. 2005; Baker et al. 2006; García et al. 2006; Hammett et al. 2006, 2012; Hutchinson et al. 2006; Keating et al. 2006; Shargie et al. 2006; Sweat et al. 2006; Decker and Montagu 2007; Hoke et al. 2007; Lönnroth et al. 2007; Noor et al. 2007; Plautz and Meekers 2007; Ross et al. 2007; Sun et al. 2007; Thuong et al. 2007; Wu et al. 2007; Garrett et al. 2008; Müller et al. 2008; O’Reilly et al. 2008; Rimal and Creel 2008; Yeung et al. 2008; Hanson et al. 2009; Hounton et al. 2009; Wang et al. 2009, 2011; Agha and Meekers 2010; Alba et al. 2010; Blanton et al. 2010; Baizhumanova et al. 2010; Casey et al. 2010, 2013; Gutierrez et al. 2010; Lutalo et al. 2010; Ngo et al. 2010; Piot et al. 2010; Qureshi 2010; Sheth et al. 2010; Angeles-Agdeppa et al. 2011; Bahromov 2011; Doyle et al. 2011; Elder et al. 2011; Hamby et al. 2011; Kassegne et al. 2011; Longfield et al. 2011; Mainkar et al. 2011; Nambiar et al. 2011; Pandey et al. 2011; Rachakulla et al. 2011; Shah et al. 2011; Thilakavathi et al. 2011; Agha and Beaudoin 2012; García et al. 2012; Harris et al. 2012; Hotz et al. 2012; Huntington et al. 2012; Patel et al. 2012; Russo et al. 2012; Azmat et al. 2013; Boily et al. 2013; Bowen 2013; Habib et al. 2013; Havemann et al. 2013; Jain et al. 2013; Juneja et al. 2013; Kang et al. 2013; Littrell et al. 2013; Loharikar et al. 2013; Monterrosa et al. 2013; Obare et al. 2013; Pawa et al. 2013; Pattanayak et al. 2009; Gupta et al. 2012). Fifteen studies contained results that were analysed for multiple health areas.
Figure 2.

Flow diagram.

Figure 2.

Flow diagram.

The greatest number of studies focused on HIV-related outcomes of social marketing, followed by reproductive health and malaria (Table 1). More studies overall came from sub-Saharan Africa than any other region (n = 67). The majority of HIV and malaria studies came from this region as well (n = 22, n = 14, respectively), while relatively few studies took place in Latin America/the Caribbean. Three studies took place in Europe/Central Asia, and three studies were multi-regional.

Table 1

Included studies by health area and region

TB HIV Reproductive Health Malaria Child Survival Total
East Asia/Pacific 10 11 32
Europe/ Central Asia
Latin America/ Caribbean 10
South Asia 10 25
Sub-Saharan Africa 22 14 14 16 67
Multi-region
Total 45 40 17 35 125a
TB HIV Reproductive Health Malaria Child Survival Total
East Asia/Pacific 10 11 32
Europe/ Central Asia
Latin America/ Caribbean 10
South Asia 10 25
Sub-Saharan Africa 22 14 14 16 67
Multi-region
Total 45 40 17 35 125a
a

Fifteen studies were analysed for multiple health areas. Total reflects the number of unique studies.

A total of 42 individual studies reported health outcomes, such or morbidity or mortality measures, in the areas of child survival, HIV, reproductive health, child survival, malaria and tuberculosis (Table 2). Baizhumanova et al. (2010) reported outcomes for both reproductive health and child survival as their intervention addressed iron deficiency anaemia. Only seven studies reporting on health outcomes used experimental study designs (Shargie et al. 2006; Ross et al. 2007; Wu et al. 2007; Lutalo et al. 2010; García et al. 2012; Habib et al. 2013). Of these, one study of reproductive health outcomes reported statistically significant decreases in occurrence of pregnancy, and one HIV/STI study reported mixed results with decreases in prevalence of hepatitis C and no change in prevalence of HIV (Wu et al. 2007; Lutalo et al. 2010). Another reproductive health study reported no difference in occurrence of pregnancy, and one HIV study reported no difference in STI prevalence (Hoke et al. 2007; Ross et al. 2007). The majority of studies used quasi-experimental designs, with HIV having the largest volume of studies followed by child survival. Almost all of the studies on child survival and malaria outcomes reported positive or mixed results. Relatively more HIV studies reported mixed results, but these studies included a range of outcomes.

Table 2

Health results by study design, direction, and statistical significance of results

Positive results

Mixed results

No significant results

Total
Experimental Quasi-experimental Experimental Quasi-experimental Experimental Quasi-experimental
HIV 13
RH
Child survival 12
Malaria
TB
Total 19 42a
Positive results

Mixed results

No significant results

Total
Experimental Quasi-experimental Experimental Quasi-experimental Experimental Quasi-experimental
HIV 13
RH
Child survival 12
Malaria
TB
Total 19 42a
a

One study reported results for both reproductive health and child survival. Total reflects the number of unique studies reporting health results.

A total of 97 unique studies across all health areas reported behavioural outcomes (Table 3). Fourteen of these 97 studies used experimental study designs. Five experimental studies reported exclusively positive results (Shargie et al 2006; Sun et al. 2007; Pattanayak et al. 2009; Lutalo et al. 2010; Hotz et al. 2012). Hotz et al. 2012 reported on both reproductive health and child survival outcomes. Four HIV RCTs reported mixed or non-significant findings (Garcia et al. 2012; Hoke et al. 2007; Wu et al. 2007; Bahromov and Weine 2011). About 85% of the included studies considering behavioural outcomes used quasi-experimental designs, with HIV leading on volume of studies. Studies on reproductive health also figured largely in the number of quasi-experimental studies measuring behaviours, with most of these studies reporting mixed findings. For example, Collumbien and Douthwaite’s (2003) evaluation of an intervention using audiotapes to disseminate contraceptive information showed statistically significant increases in use of oral contraceptives but statistically significant decreases in IUD use (Collumbien and Douthwaite 2003). Almost all of the malaria studies reported positive or mixed results. Seventeen studies reported no statistically significant findings for behaviour change, with half of the studies coming from reproductive health.

Table 3

Behavioural results by study design, direction, and statistical significance of results

Positive results

Mixed results

No significant results

Total
Experimental Quasi-experimental Experimental Quasi-experimental Experimental Quasi-experimental
HIV 20 37
RH 14 30
Child survival 14 29
Malaria 10
TB
Total 25 55 14 97a
Positive results

Mixed results

No significant results

Total
Experimental Quasi-experimental Experimental Quasi-experimental Experimental Quasi-experimental
HIV 20 37
RH 14 30
Child survival 14 29
Malaria 10
TB
Total 25 55 14 97a
a

Eleven studies reported results across multiple health areas. Total reflects the number of unique studies reporting behavioural results.

A total of 81 studies reported behavioural factor outcomes across all health areas (Table 4), and there was substantial overlap between studies that assessed behaviours and behavioural factors. In total, 63 studies reported results for both behavioural and behavioural factors. A small proportion of studies measuring behavioural factors used experimental designs (n = 8), and all of these studies reported mixed or no significant results, with more than half of these studies focused on factors related to child survival. The vast majority of studies used quasi-experimental designs, and across all health areas except TB, the greatest numbers of studies reported mixed results. For example, participants from an intervention to improve malaria treatment-seeking behaviours in Tanzania showed improvements in correctly identifying cases of malaria; however, there was no change in beliefs that malaria is preventable through bednets (Alba et al. 2010).

Table 4

Behavioural factor results by study design and direction and statistical significance of results

Positive results

Mixed results

No significant results

Total
Experimental Quasi-experimental Experimental Quasi-experimental Experimental Quasi-experimental
HIV 16 26
RH 19 28
Child survival 14 23
Malaria 15
TB
Total 18 55 12 80a
Positive results

Mixed results

No significant results

Total
Experimental Quasi-experimental Experimental Quasi-experimental Experimental Quasi-experimental
HIV 16 26
RH 19 28
Child survival 14 23
Malaria 15
TB
Total 18 55 12 80a
a

Thirteen studies reported results across multiple health areas. Total reflects the number of unique studies reporting behavioural factor results.

In total, 82 studies reported on more than one outcome type. The greatest area of overlap was in studies reporting on health behaviours and behavioural factors (n = 50), while four studies reported on health outcomes and behavioural factors (Angeles-Agdeppa et al. 2005; Baizhumanova et al. 2010; Doyle et al. 2011; Hotz et al. 2012). Seventeen studies reported on health behaviours and health outcomes. Twelve studies reported behavioural factors, behaviours, and health outcomes, with five of these studies focused on child survival outcomes, one on reproductive health, three on HIV and one on malaria.

We found 26 studies that could be scored on six or more of the Social Marketing Benchmark Criteria (Table 5). Across the 26 studies, distribution by health area was more even than that of the full analysis. The health area with the greatest number of studies was child survival (8), followed closely by HIV (7), reproductive health (6) and malaria (4). One study was on a TB intervention. One study reported results for both reproductive health and child survival.

Table 5

Results of studies meeting six or more benchmark criteria by study design, direction, and statistical significance of results

Health results

Positive results

Mixed results

No significant results

Total
Experimental Quasi-experimental Experimental Quasi-experimental Experimental Quasi-experimental
HIV
RH
Child survival
Malaria
TB
Total

Behavioural results

Positive results

Mixed results

No significant results
Total
Experimental Quasi-experimental Experimental Quasi-experimental Experimental Quasi-experimental

HIV
RH
Child survival
Malaria
TB
Total 21a
Health results

Positive results

Mixed results

No significant results

Total
Experimental Quasi-experimental Experimental Quasi-experimental Experimental Quasi-experimental
HIV
RH
Child survival
Malaria
TB
Total

Behavioural results

Positive results

Mixed results

No significant results
Total
Experimental Quasi-experimental Experimental Quasi-experimental Experimental Quasi-experimental

HIV
RH
Child survival
Malaria
TB
Total 21a
a

Two studies reported behavioural results for both reproductive health and child survival. Total reflects the number of unique studies.

The study designs and direction of results among the 26 studies were similar to that of the full analysis. Twenty-one studies reported on behavioural outcomes and 19 reported on behavioural factors—as in the full analysis, most of the 26 studies reported both outcome types together. Only five studies reported health outcomes. The majority of studies used quasi-experimental designs. Table 5 shows that for behavioural outcomes, the results were closely split between positive and mixed results; while for health outcomes, more studies reported mixed results or no statistically significant results.

### Adherence to social marketing benchmarks

We assessed how well studies adhered to the core characteristics of social marketing by investigating frequencies of benchmark criteria (Table 6). Only three studies met all 18 benchmark criteria: two described a condom campaign in urban Pakistan, and one described the social marketing of iron-fortified soy sauce in China (Sun et al. 2007; Agha and Meekers 2010; Agha and Beaudoin 2012). The most common benchmark criteria were behaviour, customer orientation, and methods mix. This suggests that most social marketing programmes directly targeted behaviour change to a specific audience of interest, using a range of communications channels to do so. Relatively few studies were coded for competition, defined as addressing ‘direct and external factors that compete for the audience's time and attention’ and attempting to decrease such competition. This may be because it is a more challenging concept to operationalize. Theory was the least commonly coded criterion, somewhat surprisingly, given that large number of studies focused on behaviour. These trends held across all outcome types. Frequencies of coded benchmark criteria were generally lower for studies reporting health outcomes. Of the studies that met six or more benchmark criteria, the most common criteria were customer orientation and insight (reported by all 26 studies), method mix and behaviour (by 25 studies each), segmentation (by 23 studies) and exchange (by 21 studies).

Table 6

Adherence to social marketing benchmark criteria

Social marketing benchmark criteria Total % (n = 125) Health outcomes % (n = 42) Behaviour outcomes % (n = 97) Behavioural factor outcomes % (n = 81)
1. Behaviour 81 79 88 75
2. Customer orientation 63 55 65 70
3. Theory 22 14 23 31
4. Insight 56 40 58 62
5. Exchange 43 40 44 41
6. Competition 26 19 29 27
7. Segmentation 42 38 41 43
8. Methods mix 84 93 84 80
Social marketing benchmark criteria Total % (n = 125) Health outcomes % (n = 42) Behaviour outcomes % (n = 97) Behavioural factor outcomes % (n = 81)
1. Behaviour 81 79 88 75
2. Customer orientation 63 55 65 70
3. Theory 22 14 23 31
4. Insight 56 40 58 62
5. Exchange 43 40 44 41
6. Competition 26 19 29 27
7. Segmentation 42 38 41 43
8. Methods mix 84 93 84 80

When we stratified studies by their categorization as having positive versus mixed results, social marketing characteristics of audience insight and exchange, i.e. addressing the costs and benefits of behaviour change, were more frequent in studies with exclusively statistically significant and positive effects than in those studies with mixed effects (insight: 80% of positive studies vs 49% mixed; exchange: 60% positive vs 39% mixed). The characteristics of competition, method mix, and audience segmentation were also found in higher frequencies in studies with positive effects than in studies with mixed results, while audience segmentation was in 42% of studies with positive results and 42% of studies with mixed results. These findings suggest that social marketing programmes designed with a deeper understanding of the target audience's motivations and fears were better equipped to influence that audience. For example, a diarrhoea-prevention social marketing programme in Thailand found through quantitative and qualitative research that cleanliness was associated with spirituality and then capitalized on this in their messaging. The evaluation found statistically significant increases in proper hand washing, dish washing, reduced fecal fingertip contamination and reduced incidence of diarrhoea (Pinfold and Horan 1996). Effective programmes also weighed costs and benefits of changing behaviour to offer appealing incentives and rewards to their target audience. Kang et al. (2013) evaluated a social marketing intervention designed to reduce HIV risk and improve treatment adherence among female sex workers in China (Kang et al. 2013). This programme aimed to reduce social barriers to HIV prevention and treatment for female sex workers through anti-stigma workshops and policy advocacy as well as increased availability and accessibility of condoms. Benefits of seeking and adhering to treatment were maximized for people living with HIV through enhanced coordination of care, including psychosocial and nutritional support, and poverty alleviation. This evaluation found statistically significant improvements in HIV knowledge and increases in condom use and receipt of HIV-related services.

We identified several evidence gaps in this review, by assessing health areas where we found less than two studies that men out inclusion criteria. In child survival, we found no studies that specifically addressed treatment of pneumonia, such as through provision of antibiotics, or socially marketing of vaccination for childhood illnesses. Within HIV, we found no studies addressing voluntary medical male circumcision, while in reproductive health we found no studies that addressed the social marketing of products for safe abortion or post-abortion care and only one study that addressed emergency contraception.

## Discussion

We aimed to assess the effectiveness of social marketing for achieving changes in health status and health behaviour in major areas of global health investment. We found evidence of the effectiveness of social marketing in global health in both health and behavioural outcomes as well as behavioural factors. Of the 125 studies meeting our inclusion criteria, roughly one-third reported on measurable health outcomes, while the bulk of the evidence assessed behavioural outcomes (n = 97) or behavioural factors (n = 81). Almost one-half of studies reported positive, statistically significant results. Effective studies acted upon audience insights and used a cost-benefit analysis to maximize incentives and increase likelihood of behaviour change. Few studies were coded as applying theories of behaviour change, suggesting some social marketing interventions included in this review did not have clear pathways to change and possibly that social marketers may need support in applying behavioural theories in intervention design.

### HIV

Much of the evidence on the effectiveness of social marketing was concentrated in HIV/AIDS, with 45 included studies. Most of these studies focused on the ability of social marketing interventions to influence condom use and other sexual behaviours. Many interventions used peer educators to disseminate HIV education and promote condoms. Studies focused on several populations including youth, female sex workers (FSW), men who have sex with men (MSM), truck drivers, migrant workers and people who inject drugs. However, only three studies considered how social marketing could influence HIV outcomes via safe injecting behaviour (Hammett et al. 2006, 2012; Wu et al. 2007). Relatively few studies addressed HIV testing behaviours (Gutierrez et al. 2010; Boily et al. 2013; Kang et al. 2013; Pawa et al. 2013). Despite being the health area with the greatest evidence base, only a small number of experimental studies were identified.

The vast majority of HIV studies were categorized as having mixed results. This may have contributed to previous reviews questioning the effectiveness of social marketing and the quality of social marketing evaluations (Pawa et al. 2013). However when considering the overall impact of social marketing interventions, many of those studies defined as ‘mixed’ had results that were mostly statistically significant and positive across outcome types. For example, in their evaluation of the Frontiers Prevention Project, which aimed to reduce STIs among men who have sex with men and female sex workers in India, Gutierrez et al. (2010) reported statistically significant decreases in prevalence of syphilis in both MSM and FSW, as well as herpes simplex virus type 2 (HSV-2) in FSW (Gutierrez et al. 2010). However, decreases found in HSV-2 among MSM were not statistically significant. Similarly, both male and female participants in the condom social marketing programme 100% Jeune in Cameroon had statistically significant increases from baseline to follow-up in ever use of condoms, use of condoms during last sex with regular and casual partners, and consistent use of condoms with casual partners (Meekers et al. 2005). Increases in consistent use of condoms with regular partners were only statistically significant among female participants. We categorized these studies as having mixed results for health and behavioural outcomes, respectively, although in both cases all results reported were positive, and only one of those results in each study was found not to be statistically significant.

### Reproductive health

Slightly less than one-third of included studies reported on reproductive health outcomes. Several HIV and reproductive health studies overlapped (Meekers 2000; Vaughn et al. 2000; Babalola et al. 2001a; Meekers et al. 2005; Kim et al. 2006; Plautz and Meekers 2007; Van Rossem and Meekers 2007; Doyle et al. 2011). These studies largely targeted adolescents with sexual and reproductive health education and promoted condoms as a means for contraception and HIV prevention. Many reproductive health studies focused on iron-folic acid supplementation to reduce and prevent iron-deficiency anaemia in women of reproductive age. Supplements were distributed for free or sold in stores. One study examined the impact of socially marketed iron-fortified wheat flour (Baizhumanova et al. 2010). Others looked at social franchising of reproductive health services, where providers received technical training such as on IUD insertion (Agha et al.2007a; Decker and Montagu 2007; Ngo et al. 2010; Qureshi 2010; Shah et al. 2011; Huntington et al. 2012; Azmat et al. 2013). Branded logos and multiple media outlets were used to promote antenatal care and contraceptive services in franchise clinics. A few interventions involved training and deployment of community health workers to generate demand for and promote family planning services.

### TB

With the smallest body of evidence, all three TB studies assessed case identification interventions. Each intervention used different methods including training of community health workers in TB outreach, education and diagnosis (Shargie et al. 2006); improved services through franchise clinics (Lönnroth et al. 2007); and promotion of screening through television and radio public service announcements (Jaramillo 2001).

### Child survival

With a smaller number of studies overall, child survival was the health area reporting the greatest proportion of exclusively statistically significant and positive results for all outcome types. Child survival studies also included a proportionately greater number of experimental studies and studies reporting health outcomes. Six child survival and reproductive health studies overlapped (Warnick et al. 2004; Kanal et al. 2005; Khan et al. 2005; Baizhumanova et al. 2010; Angeles-Agdeppa et al. 2011; Hotz et al. 2012). These six studies each focused on either iron-folic acid or vitamin supplementation for women of reproductive age, pregnant women, and/or children. Many of the included studies focused on educating caregivers on child health care or nutritional practices. For example, Havemann et al. (2013) examined the impact of a community nutrition programme that, among other methods, trained caregivers in organic farming to prevent undernutrition and stunting in children (Havemann et al. 2013). Some interventions directly targeted children with hand washing education and distribution of water treatment products in schools to prevent diarrhoeal diseases (Pinfold 1999; Curtis et al. 2001; O’Reilly et al. 2008; Blanton et al. 2010; Patel et al. 2012).

### Malaria

Malaria studies examined interventions that sold or distributed insecticide-treated nets and promoted their correct use through community dramas, text messages, music videos and community health volunteers. Several studies promoted other malaria prevention products such as DEET and insecticidal soap. Fewer studies analysed the impact of malaria treatment interventions, such as accredited drug dispensary outlets offering artemisinin-combination therapy (Alba et al. 2010). One study on community case management also reported child survival outcomes as it focused on treatment of both malaria and diarrhoea (Littrell et al. 2013).

### Evidence gaps

Although substantial evidence was found in the effectiveness of social marketing in global health, we identified evidence gaps in each health area. Male circumcision has been shown to be effective in decreasing risk of HIV acquisition (Siegfried et al. 2009; Perera et al. 2010). This evidence has led to a programmatic scale-up of voluntary medical male circumcision (VMMC) in eastern and southern Africa by USAID, the Bill & Melinda Gates Foundation, WHO, UNICEF and others, many of which programmes utilize social marketing to promote VMMC (USAID 2011). However, no evaluations assessing the effectiveness of social marketing in VMMC were identified in this review. We found three studies addressing the social marketing of tuberculosis case identification, but the evidence base in this area is still quite limited.

Due to legal limitations and political sensitivities, social marketing of medication abortion has not been implemented on as large of a scale as other reproductive health products and services (Winikoff and Sheldon 2012). However, non-governmental organizations have used social marketing to engage providers and women in access to abortion and post-abortion care in several countries. Despite this, we did not find any evidence on the effectiveness of social marketing of abortion products and services. Further, we only found one study examining social marketing of emergency contraception.

In the area of child survival, we did not find evaluations assessing child immunization interventions. Specifically, while many interventions evaluated diarrhoea prevention programs, none evaluated social marketing of rotavirus vaccines. WHO recommended scale-up of rotavirus vaccines in 2009, and programmes have since been introduced in 26 countries in sub-Saharan Africa (Mwenda et al. 2014). It is unclear as to whether social marketing techniques have been used to promote rotavirus vaccination. Additionally, no studies examined pneumonia prevention and treatment through social marketing. Pneumonia accounts for 4% of neonatal deaths and 14% of deaths in children under the age of 5 (WHO/UNICEF 2013). From this review, we cannot determine whether pneumonia has not been addressed by social marketing, or if there is a gap in evaluations of existing programs.

### Limitations

This review has several limitations. Studies were limited to those that labeled the assessed intervention as social marketing or social franchising. As such, we were not able to locate studies that did not self-identify as social marketing, despite potentially qualifying based on intervention design (Evans et al. 2014). Despite this limitation, we may also have inadvertently included studies that might be better classified as health promotion due to definitional challenges in identifying social marketing interventions. Additionally, we did not fully specify types of populations that were of interest, and we did not include grey literature, despite the likelihood of many relevant evaluations being found in the grey literature.

Our inclusion criteria included a broad range of study designs, several of which may not have been included in other review methodologies using more stringent criteria. The intent of our criteria was to identify a full range of evaluations with some degree of rigour. Other reviews have already identified the variable quality of social marketing evaluations, but for practitioners and policy-makers, a compilation of evaluations of moderate to high quality may be more valuable than relying on the highest standards of evidence, when this evidence is in short supply and not always feasible to procure (Shelton 2014). We excluded strictly observational studies.

We suspect that publication bias may have influenced the extent of positive and mixed results reported on in this review, but we did not systematically assess this. Nevertheless, we found 16 studies that reported at least one negative result. Given the range of outcomes considered in this review, we were not able to conduct a meta-analysis to provide a quantitative summary of trends.

It is possible that identified social marketing interventions were misclassified on the social marketing benchmark criteria due to minimal programme documentation included in published papers. Too often, authors focused on details of evaluation design and implementation, giving little focus to how the intervention being evaluated was designed and carried out. If social marketing interventions are evaluated to determine possibilities for scale-up and replication, greater detail on intervention design needs to be included in the peer-reviewed literature, following more detailed reporting guidelines (Quinn et al. 2010). This is especially critical for addressing risks of mis-classification between social marketing and other related intervention strategies.

### Policy implications

When we looked at the frequency of included studies by health area, HIV stands out for the numbers of included studies. Other reviews and meta-analyses have investigated the role of social marketing in HIV prevention in low and middle-income countries previously and found the results to be mixed (Bertrand et al. 2006; Noar et al. 2009; Sweat et al. 2012; Evans et al. 2014). Our results coincide with these findings, but we were able to compare HIV-related studies to those in other areas of global health investment to assess how evaluations of social marketing are distributed across health and disease areas.

We suspect that funding trends in development assistance for health explain some of how included studies were distributed by health area. Analysis of funding trends indicates that HIV has been the predominant area for global health investment for more than a decade. Maternal, newborn and child health received the second largest share of recent development assistance in analyses up to 2010 (Ravishankar et al. 2009; Murray et al. 2011). Further, HIV prevention leads in share of development assistance for women of reproductive age, with family planning receiving a smaller share (Hsu et al. 2013). Assessed according to disease burden, $330 was spent per HIV-related disability-adjusted life-year (DALY over the 2006-2010 period), while only about$75 was spent per malaria- and TB-related DALY (Hanlon et al. 2014).

We cannot determine from this review whether an evidence base on the effectiveness of social marketing follows funding trends, or if the relationship goes in the other direction. However, these findings suggest that multi-lateral and bilateral donors can certainly influence this evidence base. The potential for development assistance to invest in filling evaluation gaps has been well documented, including specific actions to be taken to ensure that budgets and timelines are structured to ensure that rigorous evidence of development effectiveness can be generated (Savedoff et al. 2006).

Specifically for social marketing, this will require investments in strengthening evaluation quality. Only a fraction of included studies in this review used experimental designs. Strengthening the quality of evidence requires investment in rigorous evaluation designs that will meet the expectations of a range of stakeholders, from those who make decisions solely based on randomized trials to those who will consider a range of methodologies calibrated to context (Shelton 2014,; Finkelstein and Taubman 2015; Hatt et al. 2015). A key part of investing in improving evaluation quality will be in ensuring that rigorous documentation of intervention strategies and processes is included in any social marketing evaluation (Quinn et al. 2010). The challenge of disentangling social marketing from other potentially effective intervention strategies such as social and behaviour change communications will continue to hamper the field of social marketing unless social marketers and their evaluators invest in documenting more explicitly how these interventions apply social marketing principles.

Another key need is to expand the evidence base on the effects of social marketing for improving health outcomes. Less than half of the studies included in this review addressed measures of morbidity, mortality or fertility. These outcomes can frequently be modelled, but models are always improved by greater access to empirical evidence, while the empirical evidence itself contributes to improving exercises in evidence synthesis (Longfield et al. 2013).

We have identified specific evidence gaps where direct evaluation investments are needed to assess the effectiveness of social marketing. These areas included childhood pneumonia, voluntary medical male circumcision, medication abortion and post-abortion care and emergency contraception. Further evaluations of the social marketing of tuberculosis treatment services are sorely needed as well. Although social marketing will certainly not be an appropriate intervention strategy in all cases, its effectiveness cannot be known until programmes currently in operation are evaluated. Better evidence is needed as well to isolate highly effective programmatic components that are scalable.

Although this review has illuminated further evidence needs, it also suggests that a sizeable evidence base exists now on whether social marketing can change behaviours and influence health outcomes in low- and middle-income countries. Assessing the content and strategy of the social marketing interventions evaluated in this review indicated that most did focus on changing specific behaviours—and not just on raising awareness, while they used a range of strategies to reach their target audiences. Audience insight and exchange were specific characteristics of programmes with positive, statistically significant results. Many of the interventions included in the review were already operating at scale. These findings improve our understanding of how to ensure that social marketing programmes can achieve measureable improvements in the lives of the people they seek to reach. As such, they provide a basis for investments to replicate and further scale effective social marketing.

## Acknowledgements

Sahana Kumar and Oana Lupu supported literature searches and data abstraction. Reshma Roshania contributed to an early conceptualization of this review. We would like to thank Kim Longfield and Olivier LeTouzé for commenting on an earlier draft of this manuscript.

### Funding

Population Services International supported the development of this systematic review.

Conflict of interest statement. None declared.

## References

Abdulla
S
Schellenberg
JA
Nathan
R
et al.  .
2001
.
Impact on malaria morbidity of a programme supplying insecticide treated nets in children aged under 2 years in Tanzania: community cross sectional study
.
British Medical Journal

322
:
270
3
.
Agha
S
Beaudoin
CE.
2012
.
Assessing a thematic condom advertising campaign on condom use in urban Pakistan
.
Journal of Health Community

17
:
601
23
.
Agha
S
Do
M
Armand
FO.
2006
.
When donor support ends: the fate of social marketing products and the markets they help create
.
Social Marketing Quarterly

12
:
28
42
.
Agha
S
Gage
A
Balal
A.
2007a
.
Changes in perceptions in quality of and access to services among clients of a fractional franchise network in Nepal
.
Journal of Biosocial Sciences

39
:
341
54
.
Agha
S
Karlyn
A
Meekers
D.
2001
.
The promotion of condom use in non-regular sexual partnerships in urban Mozambique
.
Health Policy and Plannning

16
:
144
51
.
Agha
S
Meekers
D.
2010
.
Impact of an advertising campaign on condom use in urban Pakistan
.
Studies in Family Planning

41
:
277
90
.
Agha
S
Van Rossem
R
Stallworthy
G
et al.  .
2007b
.
The impact of a hybrid social marketing intervention on inequities in access, ownership and use of insecticide-treated nets
.
Malaria Journal

6
:
13.
Agha
S.
1998
.
Sexual activity and condom use in Lusaka, Zambia
.
International Family Planning Perspectives

24
:
32
7
.
Agha
S.
2002
.
A quasi-experimental study to assess the impact of four adolescent sexual health interventions in Sub-Saharan Africa
.
International Family Planning Perspectives

28
:
67
70
.
Agha
S.
2003
.
The impact of a mass media campaign on personal risk perception, perceived self-efficacy and on other behavioural predictors
.
AIDS Care

15
:
749
62
.
Agha
S.
2011
.
Changes in the proportion of facility-based deliveries and related maternal health services among the poor in rural Jhang, Pakistan: results from a demand-side financing intervention
.
International Journal for Equity in Health

10
:
57.
Alba
S
Dillip
A
Hetzel
MW
et al.  .
2010
.
Improvements in access to malaria treatment in Tanzania following community, retail sector and health facility interventions – a user perspective
.
Malaral Journal

9
:
163.
Alisjahbana
A
Williams
C
Dharmayanti
R
et al.  .
1995
.
An integrated village maternity service to improve referral patterns in a rural area in West-Java
.
International Journal of Gynaecology and Obstetrics

48(Suppl)
:
S83
94
.
Andreasen
AR.
2002
.
Marketing social marketing in the social change marketplace
.
Journal of Public Policy and Marketing

21
:
3
13
.
Angeles-Agdeppa
I
Paulino
LS
Ramos
AC
et al.  .
2005
.
Government-industry partnership in weekly iron-folic acid supplementation for women of reproductive age in the Philippines: impact on iron status
.
Nutrtion Reviews

63(Suppl)
:
S116
25
.
Angeles-Agdeppa
I
Saises
M
Capanzana
M
et al.  .
2011
.
Pilot-scale commercialization of iron-fortified rice: effects on anemia status
.
Food and Nutrition Bulletin

32
:
3
12
.
Azmat
SK
Shaikh
BT
Hameed
W
et al.  .
2013
.
Impact of social franchising on contraceptive use when complemented by vouchers: a quasi-experimental study in rural Pakistan
.
PLOS One

8
:
e74260
.
Babalola
S
Sakolsky
N
Vondrasek
C
et al.  .
2001a
.
The impact of a community mobilization project on health-related knowledge and practices in Cameroon
.
Journal of Community Health

26
:
459
77
.
Babalola
S
Vondrasek
C
Brown
J
et al.  .
2001b
.
The impact of a regional family planning service promotion initiative in Sub-Saharan Africa: evidence from Cameroon
.
International Family Planning Perspectives

27
:
186
93
.
Bahromov
M
Weine
S.
2011
.
HIV prevention for migrants in transit: developing and testing train
.
AIDS Education and Prevention

23
:
267
80
.
Baizhumanova
A
Nishimura
A
Ito
K
et al.  .
2010
.
Effectiveness of communication campaign on iron deficiency anemia in Kyzyl-Orda region, Kazakhstan: a pilot study
.
BMC Blood Disorders

10
:
2.
Baker
EJ
Sanei
LC
Franklin
N.
2006
.
Early Initiation of and Exclusive Breastfeeding in Large-scale Community-based Programmes in Bolivia and Madagascar
.
Journal of Health, Population and Nutrition

24
:
530
9
.
Basu
I
Jana
S
Rotheram-Borus
MJ
et al.  .
2004
.
HIV prevention among sex workers in India
.
Journal of Acquired Immune Deficiency Syndromes

36
:
845
52
.
Berg
R
Mitchell
S.
2013
.
Social Marketing: Leveraging the Private Sector to Improve Contraceptive Access, Choice, and Use
.
New York
:
Futures Group
.
Bertrand
JT
O’Reilly
K
Denison
J
et al.  .
2006
.
Systematic review of the effectiveness of mass communication programs to change HIV/AIDS-related behaviors in developing countries
.
Health Education Research

21
:
567
97
.
Beyeler
N
De La Cruz
AY
Montagu
D.
2013
.
The impact of clinical social franchising on health services in low- and middle-income countries: a systematic review
.
PLoS One

8
:
e60669.
Bhutta
ZA
Ali
S
Cousens
S
et al.  .
2008
.
Interventions to address maternal, newborn, and child survival: what difference can integrated primary health care strategies make?
Lancet

372
:
972
89
.
Blanton
E
Ombeki
S
Oluoch
GO
et al.  .
2010
.
Evaluation of the role of school children in the promotion of point-of-use water treatment and handwashing in schools and households: Nyanza Province, Western Kenya, 2007
.
American Journal of Tropical Medicine and Hygiene

82
:
664
71
.
Boily
MC
Pickles
M
Lowndes
CM
et al.  .
2013
.
Positive impact of a large-scale HIV prevention programme among female sex workers and clients in South India
.
AIDS

27
:
1449
60
.
Boulay
M
Storey
JD
Sood
S.
2002
.
Indirect exposure to a family planning mass media campaign in Nepal
.
Journal of Health Communication

7
:
379
99
.
Bowen
HL.
2013
.
Impact of a mass media campaign on bed net use in Cameroon
.
Malaria Journal

12
:
36.
Bruce
J.
1990
.
Fundamental elements of the quality of care: a simple framework
.
Studies in Family Planning

21
:
61
91
.
Buchanan
DR
Reddy
S
Hossain
Z.
1994
.
Social marketing: a critical appraisal
.
Health Promotion International

9
:
49
57
.
Cairns
G
Mackay
B
MacDonald
L.
2011
. Social marketing and international development. In
Hastings
G
Angus
K
Bryant
C
(eds).
The SAGE Handbook of Social Marketing
.
London
:
SAGE
.
Casey
GJ
Jolley
D
Phuc
TQ
et al.  .
2010
.
Long-term weekly iron-folic acid and de-worming is associated with stabilised haemoglobin and increasing iron stores in non-pregnant women in Vietnam
.
PLoS One

5
: e15691.
Casey
GJ
Montresor
A
Cavalli-Sforza
LT.
2013
.
Elimination of iron deficiency anemia and soil transmitted helminth infection: evidence from a fifty-four month iron-folic acid and de-worming program
.
PLoS Neglected Tropical Diseases

7
:
e2146
.
Cheng
H
Kotler
P
Lee
N.
2011
.
Social Marketing for Public Health: Global Trends and Success Stories
.
Sudbury, MA
:
Jones & Bartlett
.
Collumbien
M
Douthwaite
M.
2003
.
Pills, injections, and audiotapes: reaching couples in Pakistan
.
Journal of Biosocial Science

35
:
41
58
.
Crape
BL
Kenefick
E
Cavalli-Sforza
T
et al.  .
2005
.
Positive impact of a weekly iron-folic acid supplement delivered with social marketing to Cambodian women: compliance, participation, and hemoglobin levels increase with higher socioeconomic status
.
Nutrition Reviews

63
:
S134
8
.
Curtis
V
Kanki
B
Cousens
S
et al.  .
2001
.
Evidence of behaviour change following a hygiene promotion programme in Burkina Faso
.
Bulletin of the World Health Organization

79
:
518
27
.
De Pee
S
Bloem
MW
Satoto
et al.  .
1998
.
Impact of a social marketing campaign promoting dark-green leafy vegetables and eggs in central Java, Indonesia
.
International Journal for Vitamin and Nutrition Research

68
:
389
98
.
Decker
M
Montagu
D.
2007
.
Reaching youth through franchise clinics: assessment of KENYAN private sector involvement in youth services
.

40
:
280
2
.
Doyle
AM
Weiss
HA
Maganja
K
et al.  .
2011
.
The long-term impact of the MEMA kwa Vijana adolescent sexual and reproductive health intervention: effect of dose and time since intervention exposure
.
PLoS One

6
:
e24866
.
Dunston
C
McAfee
D
Kaiser
R
et al.  .
2001
.
Collaboration, cholera, and cyclones: a project to improve point-of-use water quality in Madagascar
.
American Journal of Public Health

91
:
1574
6
.
Elder
JP
Botwe
AA
Selby
RA
et al.  .
2011
.
Community trial of insecticide-treated bed net use promotion in southern Ghana: the Net Use Intervention study
.
Translational Behavioral Medicine

1
:
341
9
.
Evans
WD
Pattanayak
SK
Young
S
et al.  .
2014
.
Social marketing of water and sanitation products: a systematic review of peer-reviewed literature
.
Social Science and Medicine

110
:
18
25
.
Family Planning 2020
.
2014
.
Partnership in Progress 2013-2014
.
Washington, DC
:
FP2020
.
Finkelstein
A
Taubman
S.
2015
.
Randomize evaluations to improve health care delivery
.
Science

347
:
720
2
.
Ford
N
Koetsawang
S.
1999
.
A pragmatic intervention to promote condom use by female sex workers in Thailand
.
Bulletin of the World Health Organization

77
:
888
94
.
García
PJ
Holmes
KK
Cárcamo
CP
et al.  .
2012
.
Prevention of sexually transmitted infections in urban communities (Peru PREVEN): a multicomponent community-randomised controlled trial
.
Lancet

379
:
1120
8
.
García
SG
Lara
D
Landis
SH
et al.  .
2006
.
Emergency contraception in HONDURAS: knowledge, attitudes, and practice among urban family planning clients
.
Studies in Family Planning

37
:
187
96
.
Garrett
V
Ogutu
P
Mabonga
P
et al.  .
2008
.
Diarrhoea prevention in a high-risk rural Kenyan population through point-of-use chlorination, safe water storage, sanitation, and rainwater harvesting
.
Epidemiology and Infection

136
:
1463
71
.
Glanz
K
Rimer
BK
Lewis
FM.
2002
. The scope of health behavior and health education. In:
Glanz
K
Rimer
BK
Lewis
FM
(eds).
Health Behavior and Health Education: Theory, Research, and Practice
.
San Francisco, CA
:
Jossey-Bass
,
3
21
.
Goldstein
S
Usdin
S
Scheepers
E
et al.  .
2005
.
Communicating HIV and AIDS, what works? A report on the impact evaluation of Soul City’s fourth series
.
Journal of Health Communication: International Perspectives

10
:
465
83
.
Gordon
R
McDermott
L
M
et al.  .
2006
.
The effectiveness of social marketing interventions for health improvement: what’s the evidence?
Public Health

120
:
1133
9
.
Gupta
N
Mutukkanu
T
A
et al.  .
2012
.
Preventing waterborne diseases: analysis of a community health worker program in Rural Tamil Nadu, India
.
Journal of Community Health

37
:
513
9
.
Gutierrez
JP
McPherson
S
Fakoya
A
et al.  .
2010
.
Community-based prevention leads to an increase in condom use and a reduction in sexually transmitted infections (STIs) among men who have sex with men (MSM) and female sex workers (FSW): the Frontiers Prevention Project (FPP) evaluation results
.
BMC Public Health

10
:
497.
Guttman
N
Salmon
CT.
2004
.
Guilt, fear, stigma and knowledge gaps: ethical issues in public health communication interventions
.
Bioethics

18
:
531
52
.
Habib
MA
Soofi
S
K
et al.  .
2013
.
A study to evaluate the acceptability, feasibility and impact of packaged interventions (Diarrhea Pack) for prevention and treatment of childhood diarrhea in rural Pakistan
.
BMC Public Health

13
:
922.
Hamby
A
Pierce
M
Daniloski
K
et al.  .
2011
.
The use of participatory action research to create a positive youth development program
.
Social Marketing Quarterly

17
:
2
17
.
Hammett
TM
Jarlais
DC
Kling
R
et al.  .
2012
.
Controlling HIV epidemics among injection drug users: eight years of cross-border HIV prevention interventions in Vietnam and China
.
PLoS One

7
:
e43141
.
Hammett
TM
Kling
R
Johnston
P
et al.  .
2006
.
Patterns of HIV prevalence and HIV risk behaviors among injection drug users prior to and 24 months following implementation of cross-border HIV prevention interventions in Northern Vietnam and Southern China
.
AIDS Education and Prevention

18
:
97
115
.
Hanlon
M
Graves
CM
Brooks
BPC
et al.  .
2014
.
Regional variation in the allocation of development assistance for health
.
Globalization and Health

10
:
8.
Hanson
K
Marchant
T
Nathan
R
et al.  .
2009
.
Household ownership and use of insecticide treated nets among target groups after implementation of a national voucher programme in the United Republic of Tanzania: plausibility study using three annual cross sectional household surveys
.
British Medical Journal

339
:
b2434.
Harris
JR
Patel
MK
Juliao
P
et al.  .
2012
.
Addressing inequities in access to health products through the use of social marketing, community mobilization, and local entrepreneurs in Rural Western Kenya
.
International Journal of Population Research

2012
:
470598.
Hatt
LE
Chatterji
M
Miles
L
et al.  .
2015
.
A false dichotomy: RCTs and their contributions to evidence-based public health
.
Global Health: Science and Practice

3
:
138
40
.
Havemann
K
Pridmore
P
Tomkins
A
et al.  .
2013
.
What works and why? Evaluation of a community nutrition programme in Kenya
.
Public Health Nutrition

16
:
1614
21
.
Higgins
JPT
Green
S.
Cochrane Handbook for Systematic Reviews of Interventions, Version 5.1.0. The Cochrane Collaboration. http://handbook.cochrane.org/, accessed 18 March
2015
.
Hoke
TH
Feldblum
PJ
Van Damme
K
et al.  .
2007
.
Randomised controlled trial of alternative male and female condom promotion strategies targeting sex workers in Madagascar
.
Sexually Transmitted Infections

83
:
448
53
.
Hornik
R.
2002
.
Public Health Communication: Evidence for Behavior Change
.
London
:
Lawrence Erlbaum Associates
.
Hotz
C
Loechl
C
De Brauw
A
et al.  .
2012
.
A large-scale intervention to introduce orange sweet potato in rural Mozambique increases vitamin A intakes among children and women
.
British Journal of Nutrition

108
:
163
76
.
Hounton
S
Byass
P
Brahima
B.
2009
.
Towards reduction of maternal and perinatal mortality in rural Burkina Faso: communities are not empty vessels
.
Global Health Action

7
:
2
.
Hsu
J
Berman
P
Mills
A.
2013
.
Reproductive health priorities: evidence from a resource tracking analysis of official development assistance in 2009 and 2010
.
Lancet

381
:
1772
82
.
Huntington
D
Mundy
G
Hom
NM
et al.  .
2012
.
Physicians in private practice: reasons for being a social franchise member
.
Health Research Policy and Systems

10
:
25.
Hutchinson
P
Lance
P
Guilkey
DK
et al.  .
2006
.
Measuring the cost-effectiveness of a national health communication program in rural Bangladesh
.
Journal of Health Communication: International Perspectives

11(Suppl)
:
91
121
.
Institute for Health Metrics and Evaluation
.
2014
.
Financing Global Health 2011: Continued Growth as MDG Deadline Approaches
.
Seattle, WA
:
IHME
.
Jacobs
B
Kambugu
FSK
Whitworth
JAG
et al.  .
2003
.
Social marketing of pre-packaged treatment for men with urethral discharge (Clear Seven) in Uganda
.
International Journal of STD and AIDS

14
:
216
221
.
Jain
A
Nuankaew
R
Mongkholwiboolphol
N
et al.  .
2013
.
Community-based interventions that work to reduce HIV stigma and discrimination: results of an evaluation study in Thailand
.
Journal of the International AIDS Society

16(Suppl 2)
:
18711
.
Jaramillo
E.
2001
.
The impact of media-based health education on tuberculosis diagnosis in Cali, Colombia
.
Health Policy and Planning

16
:
68
73
Joint United Nations Programme on HIV/AIDS (UNAIDS)
.
2012
.
Investing for Results. Results for People.

Geneva
:
UNAIDS
.
Juneja
S
Rao Tirumalasetti
V
Mishra
RM
et al.  .
2013
.
Impact of an HIV prevention intervention on condom use among long distance truckers in India
.
AIDS and Behavior

17
:
1040
51
.
Kanal
K
Busch-Hallen
J
Cavalli-Sforza
T
et al.  .
2005
.
Weekly iron-folic acid supplements to prevent anemia among Cambodian women in three settings: process and outcomes of social marketing and community mobilization
.
Nutrition Reviews

63
:
S126
33
.
Kang
D
Tao
X
Liao
M
et al.  .
2013
.
An integrated individual, community, and structural intervention to reduce HIV/STI risks among female sex workers in China
.
BMC Public Health

13
:
717.
Karlyn
AS.
2001
.
The Impact Of A Targeted Radio Campaign To Prevent STIs And HIV/AIDS In Mozambique
.
AIDS Education and Prevention

13
:
438
51
.
Kassegne
S
Kays
MB
Nzohabonayo
J.
2011
.
Evaluation of a social marketing intervention promoting oral rehydration salts in Burundi
.
BMC Public Health

11
:
155.
Keating
J
Meekers
D
A.
2006
.
Assessing effects of a media campaign on HIV/AIDS awareness and prevention in Nigeria: results from the VISION Project
.
BMC Public Health

6
:
123.
Khan
NC
Hoang Thi Kim
T
Berger
J
et al.  .
2005
.
Community mobilization and social marketing to promote weekly iron-folic acid supplementation: a new approach toward controlling anemia among women of reproductive age in Vietnam
.
Nutrition Reviews

63(Suppl 2)
:
S87
94
.
Kikumbih
N
Hanson
K
Mills
A
et al.  .
2005
.
The economics of social marketing: The case of mosquito nets in Tanzania
.
Social Science and Medicine

60
:
369
81
.
Kim
YM
Bazant
E
Storey
JD.
2006
.
Smart patient, smart community: improving client participation in family planning consultations through a community education and mass-media program in Indonesia
.
International Quarterly of Community Health Education

26
:
247
70
.
Kim
YM
Kols
A
Ronika
N
et al.  .
2001
.
Promoting sexual responsibility among young people in Zimbabwe
.
International Family Planning Perspectives

27
:
11
19
.
Laukamm-Josten
U
Mwizarubi
BK
Outwater
A
et al.  .
2000
.
Preventing HIV infection through peer education and condom promotion among truck drivers and their sexual partners in Tanzania, 1990-1993
.
AIDS Care

12
:
27
40
.
Lee
NR
Kotler
PA.
2011
.
Social Marketing: Influencing Behaviors for Good
.
Thousand Oaks, CA
:
Sage
.
Liberati
A
Altman
DG
Tetzlaff
J
et al.  .
2009
.
The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration
.
PLoS Medicine

6
:
e1000100.
Littrell
M
Moukam
LV
Libite
R
et al.  .
2013
.
Narrowing the treatment gap with equitable access: mid-term outcomes of a community case management program in Cameroon
.
Health Policy and Planning

28
:
705
16
.
Loharikar
A
Russo
E
Sheth
A
et al.  .
2013
.
Long-term impact of integration of household water treatment and hygiene promotion with antenatal services on maternal water treatment and hygiene practices in Malawi
.
American Journal of Tropical Medicine and Hygiene

88
:
267
74
.
Longfield
K
Panyanouvong
X
Chen
J
et al.  .
2011
.
Increasing safer sexual behavior among Lao kathoy through an integrated social marketing approach
.
BMC Public Health

11
:
872.
Longfield
K
Smith
B
Gray
R
et al.  .
2013
.
Putting health metrics into practice: using the disability-adjusted life year for strategic decision making
.
BMC Public Health

13(Suppl 2)
:
S2.
Lönnroth
K
Aung
T
Maung
W
et al.  .
2007
.
Social franchising of TB care through private GPs in Myanmar: an assessment of treatment results, access, equity and financial protection
.
Health Policy and Planning

22
:
156
66
.
Lutalo
T
Kigozi
G
Kimera
E
et al.  .
2010
.
A randomized community trial of enhanced family planning outreach in Rakai, Uganda
.
Studies in Family Planning

41
:
55
60
.
Mah
MW
Tam
YC
Deshpande
S.
2008
.
Social marketing analysis of 20 years of hand hygiene promotion
.
Infection Control and Hospital Epidemiology

29
:
262
70
.
Mainkar
MM
Pardeshi
DB
Dale
J
et al.  .
2011
.
Targeted interventions of the Avahan program and their association with intermediate outcomes among female sex workers in Maharashtra, India
.
BMC Public Health

11(Suppl 6)
:
S2.
Mathanga
DP
Campbell
CH
Taylor
TE
et al.  .
2005
.
Reduction of childhood malaria by social marketing of insecticide-treated nets: a case-control study of effectiveness in Malawi
.
American Journal of Tropical Medicine and Hygiene

73
:
622
5
.
McDermott
L
M
Hastings
G.
2005
.
What is and what is not social marketing: the challenge of reviewing the evidence
.
Journal of Marketing Management

21
:
545
53
.
J
Pollard
R
Wheeler
M.
2003
.
Review of DFID Approach to Social Marketing
.
London
:
DFID Health Systems Resource Centre
.
Meekers
D
Agha
S
Klein
M.
2005
.
The impact on condom use of the “100% Jeune” social marketing program in Cameroon
.

36
:
530.
Meekers
D.
2000
.
The effectiveness of targeted social marketing to promote adolescent reproductive health: the case of Soweto, South Africa
.
Journal of HIV/AIDS Prevention and Education for Adolescents and Children

3
:
73
92
.
Monterrosa
EC
Frongillo
EA
Gonzalez de Cossío
T
et al.  .
2013
.
Scripted messages delivered by nurses and radio changed beliefs, attitudes, intentions, and behaviors regarding infant and young child feeding in Mexico
.
Journal of Nutrition

143
:
915
22
.
Müller
O
Allegri
MD
Becher
H
et al.  .
2008
.
Distribution systems of insecticide-treated bed nets for malaria control in Rural Burkina Faso: Cluster-Randomized Controlled Trial
.
PLoS One

3
:
e3182
.
Murray
CJL
Anderson
B
Burstein
R
et al.  .
2011
.
Development assistance for health: trends and prospects
.
Lancet

378
:
8
10
.
Mwenda
J
Tate
J
Duncan
S
et al.  .
2014
.
Preparing for the scale-up of rotavirus vaccine introduction in Africa: establishing surveillance platforms to monitor disease burden and vaccine impact
.
The Pediatric Infectious Disease Journal

33(Suppl 1)
:
S1
5
.
Nambiar
D
Ramakrishnan
V
Kumar
P
et al.  .
2011
.
Knowledge, stigma, and behavioral outcomes among antiretroviral therapy patients exposed to Nalamdana’s Radio and Theater Program in Tamil Nadu, India
.
AIDS Education and Prevention

23
:
351
66
.
Nathan
R
Masanja
H
Mshinda
H
et al.  .
2004
.
Mosquito nets and the poor: can social marketing redress inequities in access?
Tropical Medicine and International Health

9
:
1121
8
.
Naugle
DA
Hornik
RC.
2014
.
Systematic review of the effectiveness of mass media interventions for child survival in low- and middle-income countries
.
Journal of Health Communication: International Perspectives

19
:
190
215
.
Ngo
Alden
DL
Pham
V
et al.  .
2010
.
The impact of social franchising on the use of reproductive health and family planning services at public commune health stations in Vietnam
.
BMC Health Services Research

10
:
54.
Noar
SM
Palmgreen
P
Chabot
M
et al.  .
2009
.
A 10-year systematic review of HIV/AIDS mass communication campaigns: have we made progress?
Journal of Health Communication

14
:
15
42
.
Noor
AM
Amin
AA
Akhwale
WS
et al.  .
2007
.
Increasing coverage and decreasing inequity in insecticide-treated bed net use among Rural Kenyan children
.
PLoS Medicine

4
:
e255
.
O’Reilly
C
Freeman
M
Ravani
M
et al.  .
2008
.
The impact of a school-based safe water and hygiene programme on knowledge and practices of students and their parents: Nyanza Province, western Kenya, 2006
.
Epidemiology and Infection

136
:
80
91
.
Obare
F
Warren
C
Njuki
R
et al.  .
2013
.
Community-level impact of the reproductive health vouchers programme on service utilization in Kenya
.
Health Policy and Planning

28
:
165
75
.
Pandey
A
Mishra
RM
Sahu
D
et al.  .
2011
.
Heading towards the Safer Highways: an assessment of the Avahan prevention programme among long distance truck drivers in India
.
BMC Public Health

11(Suppl 6)
:
S15
Partnership for Maternal, Newborn, & Child Health
.
2011
.
2012-15 PMNCH Strategic Framework
.
Geneva
:
WHO
.
Patel
MK
Harris
JR
Juliao
P
et al.  .
2012
.
Impact of a hygiene curriculum and the installation of simple handwashing and drinking water stations in Rural Kenyan primary schools on student health and hygiene practices
.
American Journal of Tropical Medicine and Hygiene

87
:
594
601
.
Pattanayak
SK
Yang
JC
Dickinson
KL
et al.  .
2009
.
Shame or subsidy revisited: social mobilization for sanitation in Orissa, India
.
Bulletin of the World Health Organization

87
:
580
7
.
Paulino
LS
Angeles-Agdeppa
I
Unita Mari
ME
et al.  .
2005
.
Weekly iron-folic acid supplementation to improve iron status and prevent pregnancy anemia in Filipino women of reproductive age: the Philippine experience through government and private partnership
.
Nutrition Reviews

63
:
S109
15
.
Pawa
D
Firestone
R
Ratchasi
S
et al.  .
2013
.
Reducing HIV risk among transgender women in Thailand: a quasi-experimental evaluation of the sisters program
.
PLoS One

8
:
e77113
Perera
C
Bridgewater
F
Thavaneswaran
P
et al.  .
2010
.
Safety and efficacy of nontherapeutic male circumcision: a systematic review
.
The Annals of Family Medicine

8
:
64
72
.
Peters
DH
Mirchandani
GG
Hansen
PM.
2004
.
Strategies for engaging the private sector in sexual and reproductive health: how effective are they?
Health Policy and Planning

19(Suppl 1)
:
i5
i21
.
Pinfold
JV
Horan
NJ.
1996
.
Measuring the effect of a hygiene behaviour intervention by indicators of behaviour and diarrhoeal disease
.
Transactions of the Royal Society of Tropical Medicine and Hygiene

90
:
366
71
.
Pinfold
JV.
1999
.
Analysis of different communication channels for promoting hygiene behaviour
.
Health Education Research

14
:
629
639
.
Piot
B
Mukherjee
A
Navin
D
et al.  .
2010
.
Lot quality assurance sampling for monitoring coverage and quality of a targeted condom social marketing programme in traditional and non-traditional outlets in India
.
Sexually Transmitted Infections

86(Suppl 1)
:
i56
61
.
Plautz
A
Meekers
D.
2007
.
Evaluation of the reach and impact of the 100% Jeune youth social marketing program in Cameroon: findings from three cross-sectional surveys
.
Reproductive Health

4
:
1.
Population Services International
.
2004
. PSI Behavior Change Framework “Bubbles”: Proposed Revision. http://www.psi.org/publication/psi-behavior-change-framework-revision/, accessed 29 April
2016
.
Price
N.
2001
.
The performance of social marketing in reaching the poor and vulnerable in AIDS control programmes
.
Health Policy and Planning

16
:
231
9
.
Quinn
G
Ellery
J
Thomas
K
et al.  .
2010
.
Developing a common language for using social marketing: an analysis of public health literature
.
Health Marketing Quarterly

27
:
334
53
.
Qureshi
AM.
2010
.
Case study: does training of private networks of family planning clinicians in urban Pakistan affect service utilization?
BMC International Health and Human Rights

10
:
26.
Rachakulla
HK
Kodavalla
V
Rajkumar
H
et al.  .
2011
.
Condom use and prevalence of syphilis and HIV among female sex workers in Andhra Pradesh, India - following a large-scale HIV prevention intervention
.
BMC Public Health

11(Suppl 6)
:
S1
.
Raphael
D.
2000
.
The question of evidence in health promotion
.
Health Promotion International

15
:
355
67
.
Ravishankar
N
Gubbins
P
Cooley
RJ
et al.  .
2009
.
Financing of global health: tracking development assistance for health from 1990 to 2007
.
Lancet

373
:
2113
24
.
Rimal
RN
Creel
AH.
2008
.
Applying social marketing principles to understand the effects of the Radio Diaries program in reducing HIV/AIDS stigma in Malawi
.
Health Marketing Quarterly

25
:
119
46
.
Ross
DA
Changalucha
J
Obasi
AI
et al.  .
2007
.
Biological and behavioural impact of an adolescent sexual health intervention in Tanzania: a community-randomized trial
.
AIDS

21
:
1943
55
.
Rowland
M
Freeman
T
Downey
G
et al.  .
2004
.
DEET mosquito repellent sold through social marketing provides personal protection against malaria in an area of all-night mosquito biting and partial coverage of insecticide-treated nets: a case-control study of effectiveness
.
Tropical Medicine and International Health

9
:
343
50
.
Rowland
M
Webster
J
Saleh
P
et al.  .
2002
.
Prevention of malaria in Afghanistan through social marketing of insecticide-treated nets: evaluation of coverage and effectiveness by cross-sectional surveys and passive surveillance
.
Tropical Medicine and International Health

7
:
813
22
.
Russo
ET
Sheth
A
Menon
M
et al.  .
2012
.
Water treatment and handwashing behaviors among non-pregnant friends and relatives of participants in an antenatal hygiene promotion program in Malawi
.
American Journal of Tropical Medicine and Hygiene

86
:
860
5
.
Ryan
R.
2013
. Cochrane Consumers and Communication Review Group: meta-analysis. Cochrane Consumers and Communication Review Group. http://cccrg.cochrane.org/sites/cccrg.cochrane.org/files/uploads/Meta-analysis_1.pdf, accessed 18 March
2015
.
Savedoff
W
Levine
R
Birdsall
N.
2006
.
When Will We Ever Learn? Improving Lives Through Impact Evaluation
.
Washington, DC
:
Center for Global Development Evaluation Gap Working Group
.
Schellenberg
JRMA
Abdulla
S
Minja
H
et al.  .
1999
.
KINET: a social marketing programme of treated nets and net treatment for malaria control in Tanzania, with evaluation of child health and long-term survival
.
Transactions of the Royal Society of Tropical Medicine and Hygiene

93
:
225
31
.
Schellenberg
JRMA
Abdulla
S
Nathan
R
et al.  .
2001
.
Effect of large-scale social marketing of insecticide-treated nets on child survival in rural Tanzania
.
Lancet

357
:
1241
7
.
Schopper
D
Doussantousse
S
Ayiga
N
et al.  .
1995
.
Village-based AIDS prevention in a rural district in Uganda
.
Health Policy and Planning

10
:
171
80
.
WR
Cook
TD
Campbell
DT.
2002
.
Experimental and Quasi-Experimental Designs for Generalized Causal Inference
.
Boston
:
Houghton-Mifflin
.
Shah
NM
Wang
W
Bishai
DM.
2011
.
Comparing private sector family planning services to government and NGO services in Ethiopia and Pakistan: how do social franchises compare across quality, equity and cost?
Health Policy and Planning

26(Suppl 1)
:
i63
71
.
Shargie
EB
Mørkve
O
Lindtjørn
B.
2006
.
Tuberculosis case-finding through a village outreach programme in a rural setting in southern Ethiopia: community randomized trial
.
Bulletin of the World Health Organization

84
:
112
9
.
Shefner-Rogers
CL
Sood
S.
2004
.
Involving husbands in safe motherhood: effects of the SUAMI SIAGA campaign in Indonesia
.
Journal of Health Communication

9
:
233
58
.
Shelton
JD.
2014
.
Evidence-based public health: not only whether it works, but how it can be made to work practicably at scale
.
Global Health: Science and Practice

2
:
253
8
.
Sheth
AN
Russo
ET
Menon
M
et al.  .
2010
.
Impact of the integration of water treatment and handwashing incentives with antenatal services on hygiene practices of pregnant women in Malawi
.
American Journal of Tropical Medicine and Hygiene

83
:
1315
21
.
Siegfried
N
Muller
M
Deeks
JJ
et al.  .
2009
.
Male circumcision for prevention of heterosexual acquisition of HIV in men
.
Cochrane Database of Systematic Reviews

15
:
CD003362
.
Smith
B
Schneider
KH.
2009
.
What is so special about Social Marketing? I mean, really
.
Social Marketing Quarterly

15
:
127
33
.
Spotswood
F
French
J
Tapp
A
et al.  .
2012
.
Some reasonable but uncomfortable questions about social marketing
.
Journal of Social Marketing

2
:
163
75
.
M
Gordon
R
Angus
K
et al.  .
2007
.
A systematic review of social marketing effectiveness
.
Health Education

107
:
126
91
.
Stenberg
K
Axelson
H
Sheehan
P
et al.  .
2014
.
Advancing social and economic development by investing in women’s and children’s health: a new Global Investment Framework
.
Lancet

383
:
1333
54
.
Sun
X
Guo
Y
Wang
S
et al.  .
2007
.
Social marketing improved the consumption of iron-fortified soy sauce among women in China
.
Journal of Nutrition Education and Behavior

39
:
302
10
.
Sweat
M
Kerrigan
D
Moreno
L
et al.  .
2006
.
Cost-effectiveness of environmental-structural communication interventions for HIV prevention in the female sex industry in the Dominican Republic
.
Journal of Health Communication

11
:
123
42
.
Sweat
MD
Denison
J
Kennedy
C
et al.  .
2012
.
Effects of condom social marketing on condom use in developing countries: a systematic review and meta-analysis, 1990–2010
.
Bulletin of the World Health Organization

90
:
613
22a
.
Tambashe
BO
Speizer
IS
Agbessi
A
et al.  .
2003
.
Evaluation of the PSAMAO “Roulez Protegée” mass media campaign in Burkina Faso
.
AIDS Education and Prevention

15
:
33
48
.
The Global Fund to Fight AIDS, Tuberculosis, and Malaria
.
2011
.
The Global Fund Strategy 2012-2016: Investing for Impact
.
Geneva
:
GFATM
.
The National Social Marketing Centre. 2012. Social Marketing Benchmark Criteria
. http://www.thensmc.com/sites/default/files/benchmark-criteria-090910.pdf, accessed 18 March
2015
.
Thevos
AK
Olsen
SJ
Rangel
JM
et al.  .
2002
.
Social marketing and motivational interviewing as community interventions for safe water behaviors: follow-up surveys in Zambia
.
International Quarterly of Community Health Education

21
:
51
65
.
Thilakavathi
S
Boopathi
K
Girish Kumar
C
et al.  .
2011
.
Assessment of the scale, coverage and outcomes of the Avahan HIV prevention program for female sex workers in Tamil Nadu, India: is there evidence of an effect?
BMC Public Health

11(Suppl 6)
:
S3.
Thuong
NV
Van Nghia
K
Hau
TP
et al.  .
2007
.
Impact of a community sexually transmitted infection/HIV intervention project on female sex workers in five border provinces of Vietnam
.
Sexually Transmitted Infections

83
:
376
82
.
UN Millennium Project
.
2005
.
Investing in Development: A Practical Plan to Achieve the Millennium Development Goals: Overview
.
New York
:
UN Development Program
.
USAID
.
2011
. In it to Save Lives: Scaling Up Voluntary Medical Male Circumcision for HIV Prevention for Maximum Public Health Impact. http://blog.usaid.gov/2011/06/in-it-to-save-lives-scaling-up-voluntary-medical-male-circumcision-for-hiv-prevention-for-maximum-public-health-impact/, accessed 19 March
2015
.
Valente
TW
Saba
WP.
1998
.
Mass media and interpersonal influence in a reproductive health communication campaign in Bolivia
.
Communication Research

25
:
96
124
.
Van Rossem
R
Meekers
D.
2000
.
An evaluation of the effectiveness of targeted social marketing to promote adolescent and young adult reproductive health in Cameroon
.
AIDS Education and Prevention

12
:
383
404
.
Van Rossem
R
Meekers
D.
2007
.
The reach and impact of social marketing and reproductive health communication campaigns in Zambia
.
BMC Public Health

7
:
352.
Vaughn
PW
Regis
A
St. Catherine
E.
2000
.
Effects of an entertainment-education radio soap opera on Family Planning and HIV prevention in St. Lucia
.
International Family Planning Perspectives

26
:
148
157
.
Wang
A
MacDonald
VM
Paudel
M
et al.  .
2011
.
National Scale-up of Zinc Promotion in Nepal: Results from a Post-project Population-based Survey
.
Jounal of Health, Population and Nutrition

29
:
207
17
.
Wang
B
Zhan
S
Sun
J
et al.  .
2009
.
Social mobilization and social marketing to promote NaFeEDTA-fortified soya sauce in an iron-deficient population through a public-private partnership
.
Public Health Nutrition

12
:
1751
9
.
Warnick
E
Dearden
KA
Slater
S
et al.  .
2004
.
Social marketing improved the use of multivitamin and mineral supplements among resource-poor women in Bolivia
.
Journal of Nutrition Education and Behavior

36
:
290
7
.
Weinreich
NK.
What is Social Marketing?
2006
. http://www.social-marketing.com/whatis.html, accessed 29 April 2016.
WHO/UNICEF
.
2013
.
Ending Preventable Child Deaths from Pneumonia and Diarrhea by 2025: The Integrated Global Action Plan for Pneumonia and Diarrhea (GAPPD)
.
New York
:
WHO/UNICEF
.
Winikoff
B
Sheldon
W.
2012
.
Use of medicines changing the face of abortion
.
International Perspectives on Sexual and Reproductive Health

38
:
164
6
.
World Bank Group
. Country and Lending Groups. http://data.worldbank.org/about/country-and-lending-groups, accessed 18 March 2016.
World Health Organization
.
2014
.
Global Reference list of 100 Core Health Indicators: Working Version 5
.
Geneva
:
WHO
.
Wu
Z
Luo
W
Sullivan
SG
et al.  .
2007
.
Evaluation of a needle social marketing strategy to control HIV among injecting drug users in China
.
AIDS

21(Suppl 8)
:
S115
22
.
Yeung
S
Van Damme
W
Socheat
D
et al.  .
2008
.
Access to artemisinin combination therapy for malaria in remote areas of Cambodia
.
Malaria Journal

7
:
96.
Zagré
NM
Delpeuch
F
Traissac
P
et al.  .
2003
.
Red palm oil as a source of vitamin A for mothers and children: impact of a pilot project in Burkina Faso
.
Public Health Nutrition

6
:
733
42
.