Diet and Physical Activity for the Prevention of Noncommunicable Diseases in Low- and Middle-Income Countries: A Systematic Policy Review

Summary

Background:
Diet-related noncommunicable diseases (NCDs) are increasing rapidly in low- and middle-income countries (LMICs) and constitute a leading cause of mortality. Although a call for global action has been resonating for years, the progress in national policy development in LMICs has not been assessed. This review of strategies to prevent NCDs in LMICs provides a benchmark against which policy response can be tracked over time.

Methods and Findings:
We reviewed how government policies in LMICs outline actions that address salt consumption, fat consumption, fruit and vegetable intake, or physical activity. A structured content analysis of national nutrition, NCDs, and health policies published between 1 January 2004 and 1 January 2013 by 140 LMIC members of the World Health Organization (WHO) was carried out. We assessed availability of policies in 83% (116/140) of the countries. NCD strategies were found in 47% (54/116) of LMICs reviewed, but only a minority proposed actions to promote healthier diets and physical activity. The coverage of policies that specifically targeted at least one of the risk factors reviewed was lower in Africa, Europe, the Americas, and the Eastern Mediterranean compared to the other two World Health Organization regions, South-East Asia and Western Pacific. Of the countries reviewed, only 12% (14/116) proposed a policy that addressed all four risk factors, and 25% (29/116) addressed only one of the risk factors reviewed. Strategies targeting the private sector were less frequently encountered than strategies targeting the general public or policy makers.

Conclusions:
This review indicates the disconnection between the burden of NCDs and national policy responses in LMICs. Policy makers urgently need to develop comprehensive and multi-stakeholder policies to improve dietary quality and physical activity.

Please see later in the article for the Editors' Summary

Introduction

Noncommunicable diseases (NCDs) are the leading cause of death globally. Of the 57 million global deaths in 2008, 36 million (63%) were due to NCDs, principally cardiovascular diseases, diabetes, cancers, and chronic respiratory diseases [1]. Mortality and morbidity data reveal the growing and disproportional impact of the epidemic in low- and middle-income countries (LMICs). Nearly 80% of the yearly NCD deaths—equivalent to 29 million people—are estimated to occur in LMICs. Without effective prevention and control, an estimated 41 million people in LMICs will die from NCDs by 2015 [2]. NCDs will evolve into a staggering economic burden over the next two decades [3].

Poor dietary quality (in particular, high salt intake, high saturated and trans-fatty acid intake, and low fruit and vegetable consumption) and insufficient physical activity are key risk factors for NCD development [4] and mortality worldwide [5], and are considered priority areas for international action [6]. The mean salt intake in most LMICs exceeds the recommended maximum intake [7]. Reducing salt intake to about 6 g/d could prevent annually about 2.5 million deaths globally [8]–[11], and a 15% reduction of salt intake over a decade in LMICs could forestall 3.1 million deaths [11],[12]. Fruit and vegetable intake is inadequate [13], and this situation contributes to 2.7 million NCD-related deaths per year. Despite evidence indicating that proper levels of physical activity are associated with a 30% reduction in the risk of ischemic heart disease, a 27% reduction in the risk of diabetes, and a 21%–25% reduction in the risk of breast and colon cancer [14],[15], approximately 3.2 million deaths each year are attributable to insufficient physical activity [6]. Physical inactivity is increasingly becoming prevalent in LMICs and already constitutes one of the leading causes of mortality [16]. There is also concern about excess intake of saturated and transfatty acids in LMICs, although large regional differences are observed [17],[18].

Preventing NCDs is not impossible [19]. Cecchini and colleagues analyzed population-based strategies to prevent NCDs in a number of LMICs with a high burden of NCDs [4]. Health information and communication strategies, fiscal measures, and regulatory measures for marketing or provision of nutrition information to children that promotes healthy eating and physical activity were found to yield substantial and cost-effective health gains, in particular in LMICs [4]. In addition, these interventions were found to be particularly effective when delivered as a multi-intervention package. Hence, it is crucial to translate the available evidence into sustainable policies in LMICs [6].

In May 2004, all WHO member states endorsed the Global Strategy on Diet, Physical Activity and Health, aiming to address NCDs through diet and physical activity [20]. Recently, a United Nations high-level meeting convened to discuss measures to prevent and control the global NCD epidemic and stressed the need to accelerate the policy response to it [21]. Monitoring this international commitment is important and can be achieved by systematic policy reviews. Previous policy reviews [22],[23], however, provided only a partial view of efforts undertaken to address NCDs, as they relied on survey data and did not consider the actual content of the policies. As policy documents are the culmination of existing social processes, they reflect the views of various stakeholders and are considered to be a reliable account of prevailing policy paradigms in a country [24]. We carried out a stocktaking exercise on national policy actions for NCD prevention in LMICs, and assessed the extent to which these address critical risk factors for NCDs, i.e., salt, fat, and fruit and vegetable intake, and physical inactivity. We focused on the existence and content of policies for the prevention of NCDs, not on their actual implementation.

Methods

Collection of Policy Documents

We searched the Internet (key words [“Nutrition” OR “NCD”] AND [“Policy” OR “Strategies” OR “Actions”]) for all publicly available national policies related to diet, nutrition, NCDs, and health in the countries classified as LMICs by the World Bank in 2011 [25]. We also searched the websites of the national ministries involved in nutrition or NCD prevention (i.e., ministries of health, sports, welfare, social affairs, or agriculture) and government portals as well as national nutrition societies. For those countries for which no policy was retrieved through the web search, an e-mail request stating the purpose of the study was sent to the respective bodies. A similar e-mail request was also sent to the WHO Regional Offices and to personal contacts of the research team. When no reply was obtained after repeated contact attempts and no reference to the existence of relevant policy documents was found during our Internet search, we classified the country as one for which we were unable to assess availability of policies. In addition to our search, we used the policy database of the WHO Regional Office for Europe [26] to assess policy availability.

Screening and Selection of Documents

The following inclusion criteria were used to include the policies in the analysis: (i) the policy is from a country classified as LMIC according to 2011 World Bank classification [25], (ii) the policy is officially approved by the national government, (iii) the policy is a publicly available document, published between 1 January 2004 and 1 January 2013, and (iv) the policy relates directly or indirectly to prevention of NCDs (Text S1). We report our findings as a systematic policy review (Text S2). Because we present our results by WHO region, we also excluded countries that were not official member states of WHO in 2011. There was no language restriction. The definitions of “policy,” “action plan,” and “program” vary broadly among the national documents. For the purpose of the present review, a broad definition of policy was used, and all national documents that included the national objectives and guidelines for action in the domain of diet and/or physical activity and/or prevention of NCDs were included. No document was excluded based on its title (e.g., “plan” versus “policy” versus “strategy”).

Data Analysis

Structured content analysis was conducted by coding the documents in NVivo 8 (QSR International). The documents were coded independently by two researchers to minimize bias induced by subjective coding. The coded documents were compared, and if coding agreement was <99% (as assessed using Kappa test agreement in NVivo), the coded text was manually reviewed for inconsistencies.

We coded all text that explicitly referred to actions aiming to (i) limit salt, (ii) modify fat intake, (iii) increase fruit and vegetable intake, or (iv) promote physical activity. Although we acknowledge that it is particularly the shift of fat consumption from saturated fats to unsaturated fats and the elimination of dietary trans-fatty acids that are critical for the prevention of NCDs, we extracted all strategies relating to dietary fat intake, such as reduction of total fat intake. The key words for coding were structured as a coding tree (Figure 1). A query was constructed for each topic in NVivo to extract all relevant text electronically. We present the results by the principle target audience of the actions, grouped into three categories: (i) general public and consumers, (ii) government, and (iii) private sector.

Results

Description of Policies

Out of the 144 countries classified by the World Bank as LMICs, four countries (Mayotte, West Bank and Gaza, the Republic of Kosovo, and American Samoa) were not official member states of WHO in 2011 and were excluded from the review. We note, however, the availability of a nutrition policy for West Bank and Gaza [27].

Of the 140 LMICs found in the six WHO regions (Africa, Europe, the Americas, South-East Asia, the Eastern Mediterranean, and the Western Pacific), we found information on the availability of policies for 83% (116/140) countries (Figure 2; Table S1). We were unable to assess the availability of policy documents for 24 countries, and in particular in the Eastern Mediterranean Region 40% (6 out of 15 countries in the region). In the European, African, Western Pacific, and South-East Asian Regions and the Region of the Americas, this proportion was 24% (5/21), 9% (4/45), 22% (4/18), 9% (1/11), and 13% (4/30), respectively. In total, 33 countries were excluded from review as they had no policy (n = 4), a policy published before 2004 (n = 19), or a policy document that was not officially endorsed (n = 3) or could not be circulated publicly (n = 3). In an additional four countries, a policy was reported to be available [26], but the full document could not be obtained.

For 29 countries, the policy document reviewed did not contain any of the NCD prevention strategies reviewed [28]–[56]. In 30 countries, policy strategies to improve dietary quality did not specify actions for any of the dietary risk factors reviewed here [28],[29],[32]–[38],[40]–[52],[55]–[62]. In the countries reviewed, strategies that addressed intake of salt, fat, or fruits and vegetables or the promotion of physical activity were found in 47% (54/116) of policies. These policies had a main focus on food or nutrition (n = 34), general health (n = 11), and, to a lesser extent, the prevention of NCDs (n = 9). In total, 36 countries had explicit actions in their policies to increase fruit and vegetable intake, 20 specified actions aimed to address dietary fat consumption, 23 specified actions to limit salt intake, and 35 specified actions to promote physical activity. Although generally low, the proportion of countries with a policy that targeted at least one risk factor was higher in South-East Asia and the Western Pacific than in Africa, Europe, the Americas, and the Eastern Mediterranean. Only 12% (14/116) of the LMICs reviewed (Bhutan, Jamaica, Mauritius, the Republic of Moldova, Malaysia, Indonesia, the Philippines, Cambodia, the former Yugoslav Republic of Macedonia, Jordan, Montenegro, Brazil, Iran, and Mongolia) proposed a package that addressed all four risk factors, and approximately 23% (27/116) of countries addressed only one of the risk factors (Figure 3).

Policy Actions to Limit Salt Intake

Only 20% (23/116) of the countries reviewed specified strategies to limit dietary salt intake, and eight policies detailed national targets to limit salt intake (Table 1). A large majority (83%; 19/23) of the countries with salt reduction strategies outlined measures of education and awareness creation in the general public and consumers, in particular focused on food labeling and promotion of foods, snacks, and packaged seasonings with reduced salt content. Strategies targeted towards the private sector were observed in 30% (7/23) of the policies and mainly related to product reformulation. The actions targeted at the governments were essentially the establishment of fiscal measures, labeling, and development of standards for salt in food and market regulations. Of the 83 countries with a policy eligible for review, 43 contained specific measures for salt iodization, but only ten of these also mentioned the need to reduce or manage dietary salt intake.

Policy Actions to Modify Fat Intake

Of the countries with strategies to modify fat intake, 65% (13/20) proposed strategies targeting the general public and consumers via public education and awareness creation (Table 2). The use of dietary guidelines and food labeling were specifically mentioned as means of public education on dietary fat intake reduction in the Mauritius [63], Bulgaria [64], Jamaica [65], and Bhutan [66]. Imposition of fiscal measures, collaboration with the food industry for product reformulation, and the establishment and enforcement of food standards were mentioned as the main actions to be implemented by the government. Only Mauritius [63], Bulgaria [64], the former Yugoslav Republic of Macedonia [67], Iran [68], and Mongolia [69] outlined specific strategies targeted towards the private sector. Intake of specific fatty acids, and in particular saturated fat (Mongolia [69], the former Yugoslav Republic of Macedonia [67], Jordan [70], Bhutan [66], Cambodia [71], Bulgaria [64], Seychelles [72], the Philippines [73], Montenegro [74], and Iran [68]) and trans-fatty acids (the former Yugoslav Republic of Macedonia [67], Bhutan [66], Bulgaria [64], Seychelles [72], the Philippines [73], Montenegro [74], Mauritius [63], Brazil [75], and Iran [68]), was addressed in ten and nine countries, respectively. Whereas Mauritius [63] proposed research into the safety of reused oils, others focused on the type of fat (Iran [68] and Cambodia [71]) or the number of times oil should be used (Seychelles [72]). Six countries (Mauritius [63], Bulgaria [64], the former Yugoslav Republic of Macedonia [67], Montenegro [74], Malaysia [76], and Viet Nam [77]) mentioned specific national fat intake targets (Table 2).

Policy Actions to Increase Fruit and Vegetable Intake

Compared to the other dietary risk factors reviewed, the objective of increasing fruit and vegetable consumption had the highest coverage: 31% (36/116) of the policies reviewed (Table S1). Promotion of school gardening, home gardening, and urban agriculture were the main actions to ensure availability and accessibility of fruit and vegetables (Table 3). The majority (75%; 27/36) of the policy documents with strategies for fruit and vegetable intake focused on public education and demonstrations to promote increased fruit and vegetable intake. Malaysia proposed the development of special recipe books in this regard [76]. Other strategies, as found in Sri Lanka [78] and Mongolia [69] for instance, targeted the catering services in educational and government institutions to ensure strict inclusion of fruits and vegetables in the meals. In all of the WHO regions, policy documents that addressed increasing fruit and vegetable consumption included the need to produce, store, and process local fruits and vegetables, and to educate populations to consume them. Policy measures outlining responsibilities for the private sector were less frequently encountered (28%; 10/36) than those detailing actions to be implemented by the government (53%; 19/36) or targeting the general public (75%; 27/36).

Policy Actions to Increase Physical Activity and Address Sedentary Lifestyle

Public education and sensitization were the main strategies to promote physical activity in the policies (Table 4). Whereas countries such as Morocco [59], Mongolia [69], and Mauritius [63] targeted educational institutions, others, such as Bhutan [66], Guyana [79], and Malaysia [76], focused on workplaces. Samoa [80], the Niger [61], Indonesia [81], India [82], and Cambodia [71] targeted the community at large. Nine countries (Kenya [60], Morocco [59], Cuba [83], Uruguay [58], Jamaica [65], Brazil [75], Malaysia [76], the Philippines [73], and China [84]) proposed national policy targets for physical activity (Table 4).

Four countries' policy documents (Georgia [57], Mongolia [69], Mauritius [63], and Chile [85]) contained detailed actions and elaborated an implementation plan for stakeholders. The need to develop sports infrastructure and urban planning (e.g., bicycle lanes and recreational centers) featured in the policy documents of Georgia [57], the Republic of Moldova [86], Turkey [62], and Mongolia [69], for instance. Five countries (Mauritius [63], Brazil [75], Samoa [80], the Republic of Moldova [86], and Serbia [87]) mentioned the need to promote physical activity among the elderly. Only four countries (Bhutan [66], the Philippines [73], Cuba [83], and the Republic of Moldova [86]) outlined specific strategies to address sedentary lifestyles, and five (Turkey [62], Cambodia [71], Jamaica [65], Serbia [87], and India [82]) documented explicit actions to involve the private sector in the promotion of physical activity.

Discussion

Despite the global disease burden of NCDs in LMICs, policies that address at least one risk factor for NCDs were found in a minority of the LMICs reviewed, and only a handful of them comprehensively tackled NCDs through integrated action on various risk factors. Even if the 24 countries with unknown existence of a NCD prevention policy actually have such a policy, the proportion with countries tackling a risk factor would amount to 56% (78/140). This finding is discouraging, because in 2004, all countries expressed a strong commitment to action to address lifestyle, diet, and physical activity [20]. Our results show that, in spite of that official commitment, most LMICs are poorly prepared to tackle the NCD increase and that little progress has been made in recent years. This finding is consistent with the results of Alwan et al. [23], who reported the results of a survey in 2010 that was limited to countries with high NCD-related mortality.

Most of the policies in our review were poorly accessible and were only obtained after an extensive search or through personal contacts. Such a situation is certainly not favorable for benchmarking and communication of policies. In agreement with Sridhar et al. [88], we argue how better sharing of best practices and lessons learned with regard to policy development is needed to address the current NCD pandemic. Additional instruments and platforms to share lessons learned in policy development and implementation are needed. Policy databases with links to documents were created previously, but are restricted to nutrition action [89] or the European region [26]. An open-access, full-text global repository of initiatives and policies to address NCDs would be a great step forward. It could also contribute to global leadership and shared accountability in the global fight against NCDs, an issue that is long overdue [90]. Ideally, such a policy database would be connected to surveillance data on the main NCD risk factors, as suggested previously [23], and would facilitate tracking progress in the coming years. We are ready to organize such an open-access repository and invite interested policy makers to contact us for an update of the current database.

Priority setting and clear articulation of what needs to be done by stakeholders is a second key issue that emerged in this analysis. Countries seasoned in the fight against NCDs develop comprehensive strategies that focus on critical risk factors and what is expected of stakeholders [91]. In the present analysis, the level of detail and outlining of the organization of policy actions to undertake was generally discouraging. Only a minority of the policies reviewed surpassed description of policy actions and included a budget, implementation plan, time frame, and devolvement of responsibility for strategies to combat specific risk factors. Various policies describe strategies and actions for NCD prevention as “the need to develop and review dietary guidelines and recommendations for people suffering from nutrition-related NCDs” or use generic statements such as “create awareness of healthy eating lifestyle to control NCDs.” Such general statements are not informative, and clear actions need to be outlined in the policies to mobilize stakeholders for effective action [92].

Since its inception during the 1992 International Conference on Nutrition [93], the approach to streamline nutrition action in national policies has had limited success, partly because of the lack of strong leadership and commitment to lead concerted action involving various stakeholders [94]. The current scientific evidence and international experience in the fight against NCDs consistently indicates the need for comprehensive and integrated action on various risk factors [95]. Mobilization of the main actors—in particular, governments, international agencies, the private sector, civil society, health professionals, and individuals—is imperative [96]. An important limitation of most policies included in the analysis is the absence of plans, mechanisms, and incentives to foster multi-stakeholder and cross-sector collaboration. The food and nonalcoholic beverage industry, for instance, can play a role in the promotion of healthier lifestyles. However, before engaging with the private sector, government agencies should be aware of the need to manage potential conflicts of interest between the government and the private sector and should try to address these by defining clear roles, responsibilities, and targets to be achieved as a result of their collaboration [97]. Most strategies encountered in the policies were directed towards government agencies and consumers, and few were targeted at the business community, international agencies, or civil society. The United Nations Political Declaration on NCDs makes a strong call for multi-stakeholder partnerships to be leveraged for effective prevention of NCDs. Policy makers in LMICs may need additional support for the development of multi-stakeholder collaborations to address the burden imposed by NCDs as well as their root causes.

In our review of governmental policies relating to NCD prevention in LMICs, strategies to increase fruit and vegetable intake were the most frequent dietary action for NCD prevention. This is hardly surprising, as fruit and vegetable interventions were taken up early on in LMICs, primarily to address prevailing micronutrient deficiencies such as vitamin A deficiency [98]. Many of these experiences, however, are restricted to the development of food-based dietary guidelines or incentives targeted towards the agricultural sector. Policy measures to achieve better diet will require constructively engaging much more with a wider range of stakeholders, in particular the food industry, retail, and the catering sector [99]. The difficulty of developing a comprehensive policy response and integrated package of strategies is not restricted to NCDs alone, and has previously been observed in an in-depth analysis of high-burden countries for child malnutrition [100]. We also note that various countries have developed strategies to reduce total fat intake, despite convincing evidence that it is the reduction of saturated and trans-fatty acids in particular, and not total fat intake, that is effective to address NCDs [101].

Most strategies encountered in the policy documents focused on consumers and aimed to prevent NCDs through awareness creation, education (i.e., labeling), or changing individuals' behavior. The traditional approach to addressing lifestyle changes in individuals has met with very limited success. It is widely accepted that the environmental context drives individual diets and lifestyle [102] and that programs need to incorporate environmental determinants (i.e., the quantity, quality, or price of dietary choices, or the built environment for physical activity) in order to be effective. Such policy measures, in particular those addressing the private sector, were poorly elaborated in the policy documents [103].

A key issue is the actual implementation of policy measures in relation to what was articulated in the documents. The findings of this review indicate that few LMICs have made significant steps in the development of a comprehensive set of strategies to address NCDs. Although an in-depth evaluation of actual implementation, effects, and resources allocated has not been opportune to date, we hope that our findings provide baseline data and encourage countries to develop monitoring and evaluation mechanisms to assess policy response in due time. Documenting the effectiveness of population-based NCD prevention policies will be a critical factor of success to ensure effective action in LMICs [4].

For this review, we were able to assess documents in all languages received. Because of language constraints, however, two of the documents [74],[87] were coded by only one researcher. To assess the content of the policy of Iran, we relied on translations by experienced senior Iranian researchers. All other policy documents were obtained in Spanish, Portuguese, French, or English and were analyzed accordingly by the research team. For China and the Russian Federation, appropriate English versions of the policies were obtained from the Chinese Centers for Disease Control and the United States Department of Agriculture, respectively. Despite indications of availability of relevant policies in the European region [26], language limitations did not allow us to search the websites of a number of countries such as Azerbaijan, Belarus, and the Russian Federation.

The restriction of our review to only national policies presents a number of limitations. The mere presence or absence of policies or strategies for NCDs in a policy document does not necessarily reflect concrete action. Conversely, nutritional interventions have been implemented in some countries without a policy being developed and published [104]. In addition, this review assessed the contents of the policy documents as they were published and did not capture local or regional activities, or initiatives that emerged after the publication of the policies. The findings from a survey in countries with a high burden of NCDs, such as Thailand and South Africa, illustrate this discrepancy [23]. The contents might have been modified over time in response to new scientific findings, emerging nutritional challenges, or changes in the countries' priorities [91]. In addition, it is important to point out that we extracted only actions that explicitly referred to one of the risk factors analyzed. Generic statements such as “development of food-based dietary guidelines” or “establishment of fiscal measures for a healthy diet” were hence not coded.

The present review shows that the policy response to address current NCD challenges through diet and physical inactivity in LMICs is inadequate since endorsement of the Global Strategy on Diet, Physical Activity and Health [20]. LMICs urgently need to scale up interventions and develop integrated policies that address various risk factors for NCD prevention through multi-stakeholder collaboration and cross-sector involvement. Clear and prioritized actions are needed to harness the NCD epidemic. Such actions need to be documented in policy documents that are publicly available to share lessons learned, promote engagement with the stakeholders, and stimulate accountability and leadership in the fight against the burden of NCDs in LMICs. The establishment of an open-access and publicly accessible database of policy documents with regular systematic reviews of policy development might prove to be an incentive in this regard.

Supporting Information

Attachment 1

Attachment 2

Attachment 3

Attachment 4

Attachment 5

Attachment 6