Community-based interventions to increase dairy intake in healthy populations: a systematic review

Background Considering the low frequency of dairy intake in the population, interventions aiming to increase its consumption can be a priority for any health system. Objective This study aims to summarize community-based interventions for improving dairy consumption and their effectiveness to help policy-makers in designing coherent public health strategies. Methods This study was conducted in 2019, using PubMed/MEDLINE, Scopus, EMBASE, Cochrane Library, Web of Science, ProQuest, and Google Scholar. Two independent reviewers selected the eligible studies, and the outcomes of interest were extracted. The quality of eligible studies was assessed using the Joanna Briggs Institute Critical Appraisal Checklist for randomized controlled trials and quasi-experimental studies. Results Out of 521 initially identified articles, 25 studies were included. Interventions reported in 19 studies were effective in increasing dairy consumption. Interventions in high-income countries were more effective than those in middle- and low-income countries. Interventions in health centers and supermarkets were more effective than the community and school-level interventions. Interventions in supermarkets and adolescents as target groups were more effective than children, middle-aged people, and the elderly. Also, educational interventions and changing buying/selling pattern were more effective than multiple interventions. Interventions longer than 24 and 48 weeks were more effective than shorter interventions. Conclusion Three policy options including educational interventions, multiple interventions, and changing the purchase pattern are suggested. It seems that applying all of the interventions together can be more effective. Also, long-term and well-designed future studies in different settings are recommended to confirm these results.


Background
Dairy is a rich source of proteins, vitamins, and minerals. It fulfills the vital needs of humans in different periods of life. So, consuming dairies has a great importance regardless of age [1]. Milk and dairy products are rich sources of calcium and magnesium [2]. These minerals are essential for the growth and strength of the bones, especially in children and adolescents due to rapid skeleton growth and bone condensation [3]. In addition, calcium is necessary for the natural mineralization of the bones and the matrix of cartilage [4]. According to the evidence, calcium can also reduce the risk of osteoporosis in the older population [5][6][7][8]. In adulthood, consuming dairy is associated with the risk reduction for some chronic diseases [9,10]. Increased intake of dairy products using a restricted diet can accelerate weight loss [11,12]. The results of a systematic review showed that an average increase in the consumption of low-fat dairy (200 g per day) reduces the risk of diabetes [13]. Also, a meta-analysis of cohort studies demonstrated that total milk and dairy consumption have an inverse relation with the incidence of colorectal cancer [14]. Since dairy products have a proper amount of calcium, magnesium, and potassium-which play an important role in regulating blood pressure-they can be useful in decreasing blood pressure and preventing stroke [15].
According to studies, dairy intake is considered low among children and adolescents [16,17]. According to the United States Department of Agriculture (USDA), the daily recommendation of dairy for an adult with an energetic need of 2000 Kcal/day is 3 portions per day, but the US population only consumes half of the recommendation (1.5 portions per day) [18]. Also, some other studies conducted in different age groups indicated that the intake was below the recommended daily amount of dairy [19][20][21][22][23][24][25][26][27][28].
Although dairy consumption has countless benefits for improving health issues, the amount of dairy intake is lower than the international recommendations [22][23][24]29], which can lead to an increase in the incidence of certain diseases such as osteoporosis, diabetes, colorectal cancer, and hypertension as well as healthcare costs [30]. In order to identify policy options to increase dairy intake, reviewing successful community-based strategies and their effectiveness is needed to help policy-makers in choosing the best policy options with respect to situations in each country. Considering the numerous evidence available and the lack of a systematic review in this field, this study was conducted to evaluate community-based interventions for improving dairy consumption and their effectiveness to help policy-makers in designing coherent public health strategies.
Science were systematically searched. The timeframe selected for searching articles was from 2000 to 2019. A number of prestigious journals in the field were also searched manually to identify and cover more articles, mainly the ones published after the database search was performed. After excluding articles which did not meet the inclusion criteria, the reference lists of the remaining articles were also searched to increase the reliability of identifying and reviewing the eligible articles.

Inclusion and exclusion criteria
Randomized clinical trials (RCTs) or quasi-experimental studies were included if they (i) enrolled all age groups of normal and healthy population, (ii) evaluated the effect of community-based intervention to increase dairy consumption, and (iii) were published in an English language journal.
Studies conducted on subjects with chronic diseases such as cardiovascular diseases, diabetes, and hypertension were excluded from the study. Also, studies which focused on calcium intake instead of dairy consumption were removed. Economic evaluations, modeling studies, laboratory and observational studies, and studies with the aim of tool development were also excluded. The research question based on PICO is available in Table 1.
Study selection, data extraction strategy, and quality assessment Two reviewers extracted the data independently and screened the title and abstract of records to identify which potentially relevant records met the inclusion/exclusion criteria. Full-text articles were obtained for these records and were independently assessed for relevance by the reviewers. The following information was extracted from the studies: author(s), year of publication, target population characteristics, country, methodological characteristics (study design), types and aims of intervention, and main outcomes. Countries were categorized as low-, middle-, and high-income according to the World Bank [32] for the current 2019 fiscal year. "Low-income economies were defined as those with a gross national income (GNI) of $995 or less in 2017, which was calculated using the World Bank Atlas method; lower middle-income economies were those with a GNI per capita between $996 and $3895; upper middle-income economies were those with a GNI per capita between $3896 and $12,055; and high-income economies were those with a GNI per capita of $12,056 or more". The quality of studies was assessed using the Joanna Briggs Institute Critical Appraisal Checklist for RCTs and Quasi-Experimental Studies. A flow chart of the literature review is shown in Fig. 1.

Results
Out of 521 results retrieved from databases and other mentioned sources, 241 were removed due to the duplicity among different sources. Reviewing the title and abstract of the results, 230 more studies were excluded due to incompatibility with the study question and aim. After reviewing the full texts of the remaining articles, 25 other studies were removed; finally, 25 articles were included in the study (Fig. 1). The data and characteristics of the 25 reviewed articles are presented in Tables 2 and 3. Out of the 25 reviewed articles, 23 were in high-income countries. Most studies (n: 16) were conducted in the USA. The studies could be divided into four categories based on the context: at the community level (n: 6), schools (n: 11), health centers (n: 5), and markets (n: 3). Also, most studies (n: 16) were randomized controlled trials. Participants in four studies were adults and the elderly. Meanwhile, children and adolescents were the point of research in 2 and 17 studies, respectively. Totally, the number of subjects in the intervention and control groups was 6939 and 6753, respectively.
Educational interventions were used in 19 studies. In addition, three studies focused on the sale and providing dairy products and three studies used multiple-level interventions. In six studies, increasing the intake of dairy products was the primary goal and in 19 studies multi-component aims were applied. Regarding the duration of interventions, a maximum of 350 weeks and a minimum of 5 weeks were performed. The intervention was repeated daily in six studies, once a week in 13 studies, once every 2 weeks in 4 studies, and every month in 1 study. In one study, the time of repeating the intervention was not mentioned [34].

Location/field of study
The majority of studies were performed at school level (n: 11), of which six studies were effective in increasing dairy intake. There were 6 studies at the community level, of which 4 were effective. There were 5 studies in health centers, all of which confirmed the effectiveness of the interventions. Fewer interventions were performed in the supermarkets (n: 3); the results of all 3 studies showed the effectiveness of the interventions.

Target group
Adolescent groups were the most common target groups (n: 17), with a total of 12 studies confirming the effectiveness of the interventions. Children were the target group in two studies, of which none were effective. The adults and older people were the target group of four studies, three of which were effective. The supermarket intervention was in three studies, all of which were effective in increasing the intake of dairy products.
Multiple interventions (n: 2) also included public advertisement and lowered prices for healthy products in schools or supermarkets; all of these interventions were effective [40,50].
Providing dairy products for children in schools [41,48], and increased exposing with the recommended products or prime placement of healthy products in markets were some other types of interventions [53].

Increasing dairy products consumption as a main goal
In six studies, increasing the intake of dairy products was the primary goal, of which four were effective (66.6%). Moreover, of the 19 studies with other goals, 14 were effective (73.6%) in increasing the intake of dairy products.

Study duration
The included studies were divided into four groups: studies with less than 8 weeks, including eight studies, of which four were effective; studies with nine to 24 weeks, including four studies, of which two were effective; studies with 24 to 48 weeks, including nine studies, of which eight were effective, and four studies with more than 48 weeks, of which all were effective.

Intervention frequency
Six studies were repeated daily (repetition due: 1 day), four of which were effective. Most studies (n: 13) had a weekly repeat interval, nine of which were effective. Four studies were repeated every 2 weeks, all of which were effective. Also, one study was repeated every 1 month, which was effective.
Interventions reported in 18 studies were effective in increasing dairy consumption and ineffective in seven studies. The effectiveness of the interventions based on the studied variables (country, location/field of study, participants, type of intervention, combination/specific dairy intervention, duration, and frequency) is shown in Table 4. In low-and middle-income countries (LMIC) and high-income countries 100% and 70% of the interventions were effective, respectively. Interventions in health centers and supermarkets were more effective than community-and school-level interventions. Interventions in adolescents as target groups were more effective than children, middleaged, and the elderly people. Also, multiple interventions and changes in buying/selling habits were more effective than educational interventions. Combined interventions to modify dairy intake were more effective than specific interventions. Interventions longer than 24 and 48 weeks were more effective than shorter interventions. Also, interventions repeated every 2 weeks and every month were more effective than interventions repeated daily or weekly.   School nutrition program including policy, education, food provision, and family and community involvement. An informative handout for parents was given to students (influencing the home environment and role models, and influencing self-efficacy). Healthy breakfast and snacks were provided (including vegetables or fruit, whole grains, protein sources, and milk or milk alternatives).

Quality assessment of studies
As can be seen in Tables 5 and 6, all studies had one or more domains characterized as high risk. Also, all studies had good quality in terms of having more low-risk domains than high-risk ones. Ten studies had selection bias due to a non-random selection of participants. Most of the studies had performance bias due to the unblinding of participants and personnel.

Discussion
Out of 521 retrieved articles, 25 were finally included in the study. The reported interventions were effective in increasing dairy consumption in 18 studies and ineffective in seven studies. As noted above, most interventions were conducted in high-income countries. This could be related to the greater research budgets. In high-income countries, different intervention methods and techniques had been used compared to lowand middle-income ones. The existing differences between these two groups of countries should be considered in applying the intervention types [58][59][60].
The results showed that the interventions performed in the healthcare centers and supermarkets were more effective than the community-and school-level interventions.
All the studies are considered as low risk of bias All the studies are considered as low risk of bias In addition, interventions in health centers are more acceptable and better adhered because of the psychological impact of these centers on participants. People who visit these centers tend to have a more positive attitude towards the role of the health system and they usually trust the recommendations of the health system [61][62][63]. Also, in community-based interventions, participants are usually adults and the elderly and changing food habits and behaviors in these groups is a really challenging task compared to children and adolescents. Therefore, it seems that designing and implementing interventions to increase dairy consumption can be better achieved if combined with the collaboration of the health system and the healthcare staff.
Regarding the effectiveness of interventions in supermarkets, all interventions in these centers were control/change interventions in the purchase of dairy products, which were much more effective. These interventions are usually made by offering discounts on dairy purchases (especially low-fat dairy) that appear to be a stronger incentive. It should be noted that a number of purchases were measured in these studies and the amount of dairy consumption was not assessed.
Based on the results, interventions on the children, the middle-aged, and the elderly people were less effective. The two studies performed in children applied multiple methods like policy changing, education, family and community involvement, and educational methods using educational CDs for children and parents. The duration of studies varied from 5 to 48 weeks. The results of review studies in different fields also indicate that the effectiveness of educational interventions on children is low [64][65][66][67][68]. The results of a systematic review and meta-analysis showed that interventions in adolescents were more effective in comparison with children [69]. Reasons for the lower effectiveness of interventions on children may be due to the low dose of the intervention [40], self-administered questionnaires which could lead children to under-or overestimate their intake [42], and lack of out-of-school nutrition programs [54].
Birch and colleagues have also stated that in order to improve primary school children's healthy food preferences, experiences and strategies are needed to increase availability and accessibility to increase exposure to those foods that will then affect their willingness to taste [70]. The low effectiveness of interventions in middle-aged and elderly people may also be attributed to the consolidation of dietary habits and other behaviors in these people; so modifying their behavior is very difficult [71,72].
The results of our study showed that most of the interventions were educational, which had lower efficiency compared with interventions such as multiple interventions and changing the purchase patterns. The results of several systematic reviews support this finding [73,74]. Although many experts believe that the effectiveness of educational interventions alone is in doubt, it seems that training can be effective if it is targeted and accompanied by other interventions to change behavior [75][76][77]. The findings of this study showed that providing dairy products is one of the effective methods to increase dairy intake. Wordell et al. demonstrated that healthful modifications in the school food environment are associated with positive food behaviors in early adolescents, but there was a cost associated with those changes. So, it seems that, if possible, providing dairy is a suitable way to increase the consumption of dairy products.
Interventions that lasted more than 24 weeks and repeated every 2 weeks and each month were more effective. The likely reason for the effectiveness of long-term interventions may be due to the effect of these interventions on the change of behavior and its stabilization. Because any short-term changes in human behavioral patterns can simply return to the basic state, but the longer-term change increases the likelihood of a sustained change in behavior [78,79]. Possible reasons for the less effectiveness of interventions that were repeated less frequently (daily and weekly) might be that participants become bored. In addition, most of the interventions that were repeated less frequently were educational and multiple interventions, that were less effective than the other interventions.
The main advantage of our systematic review is the low risk of subjective data selection. Study searches, assessment, and data synthesis were based on predefined criteria and were performed using well-established tools by two independent reviewers. Nevertheless, our analysis had some limitations. First, publication bias cannot be excluded, i.e., ineffective interventions are less likely to be published. Potential limitations existing in the included studies are as follows: unclear or inadequate allocation concealment, no intention-to-treat analysis, inadequate information about controlling confounders, and applying different questionnaires for evaluating dairy intake. One of the most important limitations of the present study is the dispersion and variability of the reported indices for the effectiveness of interventions in the studies. Hence, performing a quantitative analysis (meta-analysis) was impossible.

Conclusion
According to the results of the present systematic review, three policy options including educational interventions, multiple interventions, and changing the purchase patterns are suggested. It seems that applying all of the interventions together can be more effective. Interventions in health centers and supermarkets are more effective than the community interventions. However, it should be noted that the implementation of the proposed interventions and settings depends on the limitations, resources, and facilities of different countries. So, long-term and high-frequency interventions focusing on increasing dairy products intake are suggested in different settings and countries.