Characteristics of Multicomponent Interventions to Treat Childhood Overweight and Obesity in Extremely Cold Climates: A Systematic Review of a Randomized Controlled Trial

Aim: To analyze the characteristics of multicomponent interventions to reduce childhood overweight and obesity in territories with an extremely cold climate. Methods: A systematic review was conducted following the PRISMA statement. MEDLINE, PsycNet, SciELO, and grey literature databases were reviewed in the period between 2010 and 2020. Results: 29 articles were included (n = 4434 participants; 9.3 years; 56% women) with an average adherence of 86%, 100% being the highest adherence, for the physical activity and nutrition interventions. The primary variables studied were BMI, BMI Z-score BMI-SDS and, additionally, the secondary variables studied were nutritional status and physical and mental health. In 72% of the interventions presented, positive effects were seen on the reduction of BMI, including in parents and their children. The interventions were carried out mainly by nutritionists in health centers. The duration of the 29 interventions was ≤6 months and ≥12 months, in 59% and 41% of the studies, respectively. 57% of the studies reported post-intervention results. 86% of the interventions included a physical activity component, 80% included a nutrition component, 66% included a behavioral therapy component and 55% included an education component. Concerning the effects of the intervention on the primary outcome, in interventions with a duration equal to or less than six months, the most effective interventions included recreational activities, education, and nutritional programs. In interventions lasting 12 months or more, the most effective interventions included physical activity recommendations, nutritional and physical exercise programs, and cooking classes. Conclusions: This systematic review analyzed the effectiveness of, and characterized, multicomponent interventions lasting for 6 and 12 months, aiming to treat childhood obesity in extremely cold climates. The most frequently used units of measurement were also analyzed and summarized. Evidence derived from RCT. These results can be useful for designing future interventions to treat childhood obesity in territories with an extremely cold climate.


Introduction
The World Health Organization (WHO) has consistently reported that overweight and obesity are causing a significant deterioration in the health of the world's population [1]. These conditions have been linked for decades to the development of diseases such as

Search Strategy to Identify Studies
We reviewed the MEDLINE databases using PubMed, PsycNet, SciELO, and grey literature. The objective was to identify studies that carried out interventions for the treatment of overweight and obesity in children in extremely cold climates. The search covered the period from January 2010 to January 2020 taking as reference the Global Action Plan for the Prevention and Control of NCDs in the Americas. For this search, we used The MeSH terms: "Therapy", "Therapeutics", "Rehabilitation", "Obesity", "Overweight" and "Child". The search strategy followed the Peer Review of Electronic Search Strategies (PRESS) [25]. The search terms were: (Therapy OR Therapeutics OR Rehabilitation) AND (Obesity OR Overweight) AND (Child), adapted to each database applying the following filters: period (January 2010 to October 2020), type of article (Randomized Controlled Trial), age (6-12 years) and country (Russia, Canada, United States, Iceland, Finland, Estonia, Norway, Sweden, Denmark, Latvia, France, New Zealand, Scotland, United Kingdom, Argentina, Chile, Faroe Islands, and Iceland).

Study Selection and Inclusion Criteria
Studies that met the following inclusion criteria were selected: (a) Country: interventions carried out in countries with extremely cold climates (Russia, Canada, USA, Iceland, Finland, Estonia, Norway, Sweden, Denmark, Latvia, France, New Zealand, Scotland, United Kingdom, Argentina, Chile, Faroe Islands, and Iceland); (b) Sample: boys and girls between 6 and 12 years old; (c) Methodological Design: randomized controlled clinical trial; (d) Period: between 2010 and 2020. Exclusion criteria were: reviews, editorial documents, protocols, and doctoral thesis (Table 1). Table 1. Eligibility criteria of the studies.

Criteria Description
(1) Type of intervention: multi-component intervention Duration

Data Extraction
First, we used the Mendeley tool to look for duplicates. As no duplicates were found, then the articles that met the inclusion criteria were selected. When decisions could not be made considering only the title and abstract of the article, the full text was retrieved (Figure 1). We used a standardized questionnaire to extract the data from the included articles to synthesize the evidence. The information extracted included: (a) general characteristics of the studies and the participants (author, year, country, sample size, adherence, reasons for withdrawal, recruitment, age, sex, inclusion criteria, and place of intervention) (b) main characteristics of the interventions; (c) characteristics of the interventions according to the component addressed; (d) main variables assessed (physical health and mental health); (e) main assessment instruments used. of the studies and the participants (author, year, country, sample size, adherence, reasons for withdrawal, recruitment, age, sex, inclusion criteria, and place of intervention) (b) main characteristics of the interventions; (c) characteristics of the interventions according to the component addressed; (d) main variables assessed (physical health and mental health); (e) main assessment instruments used.

Risk of Bias Tool
To assess the methodological quality of the studies included in this review, we used the "Cochrane Handbook of Systematic Reviews of Interventions" [26], evaluating the risk of bias in each of the proposed items: (a) Selection bias, (b) Performance bias, (c) Detection bias, (d) Attrition bias and (e) Reporting bias. The only item that was not considered was "Other biases". The results of this analysis are presented, grouping all the studies analyzed ( Figure 2).

Risk of Bias Tool
To assess the methodological quality of the studies included in this review, we used the "Cochrane Handbook of Systematic Reviews of Interventions" [26], evaluating the risk of bias in each of the proposed items: (a) Selection bias, (b) Performance bias, (c) Detection bias, (d) Attrition bias and (e) Reporting bias. The only item that was not considered was "Other biases". The results of this analysis are presented, grouping all the studies analyzed ( Figure 2).

Data Synthesis Strategy
The main information of the included studies is presented in summary tables and figures. In the discussion, the most relevant methodological and applicability aspects are analyzed and some suggestions for future research are given.

Data Synthesis Strategy
The main information of the included studies is presented in summary tables and figures. In the discussion, the most relevant methodological and applicability aspects are analyzed and some suggestions for future research are given. Figure 1 shows the flow chart made following the PRISMA Statement. We identified 1056 potential studies on the treatment of childhood overweight and obesity in countries with extremely cold climates. After the exclusion of the duplicates in the databases, the screening and eligibility criteria were applied. Finally, 29 articles were included for data synthesis in this review .

Characteristics of the Studies Analyzed
29 articles were included in this review, whose interventions were carried out in the following countries considered as extreme cold climate zones: Sweden, Canada, Denmark, United States, Iceland, Germany, Norway, Switzerland, England, Finland, Holland, and China. Considering all the studies, the total number of subjects at baseline was 4434, with an average adherence of 86%. The adherence to the programs, depending on the type of the intervention, was: (1) Physical activity and nutrition: 100%; (2) education: 93%; (3) behavioral and education: 93%; (4) physical activity and education and nutrition: 87%; (5) behavioral and nutrition: 86%; (6) physical activity and behavioral: 86%; (7) physical activity and behavioral and education and nutrition: 84%; (8) physical activity and behavioral and nutrition: 81%. Of these studies, 93% reported losses in their sample, the main reasons being categorized as follows: lack of time (10%), personal reasons (24%), attendance (41%), health (20%), not meeting criteria during the development of the intervention (17%), did not wish to continue (27%) and unspecified (34%). 58% of the studies recruited children from health centers, 27% from educational centers, and 51% from the community. The mean age was 9.3 years. Concerning gender distribution, 56% were women and 44% men. Regarding the prevalence of diseases, only two studies detailed this information:  Figure 1 shows the flow chart made following the PRISMA Statement. We identified 1056 potential studies on the treatment of childhood overweight and obesity in countries with extremely cold climates. After the exclusion of the duplicates in the databases, the screening and eligibility criteria were applied. Finally, 29 articles were included for data synthesis in this review .

Characteristics of the Studies Analyzed
29 articles were included in this review, whose interventions were carried out in the following countries considered as extreme cold climate zones: Sweden, Canada, Denmark, United States, Iceland, Germany, Norway, Switzerland, England, Finland, Holland, and China. Considering all the studies, the total number of subjects at baseline was 4434, with an average adherence of 86%. The adherence to the programs, depending on the type of the intervention, was: (1) Physical activity and nutrition: 100%; (2) education: 93%; (3) behavioral and education: 93%; (4) physical activity and education and nutrition: 87%; (5) behavioral and nutrition: 86%; (6) physical activity and behavioral: 86%; (7) physical activity and behavioral and education and nutrition: 84%; (8) physical activity and behavioral and nutrition: 81%. Of these studies, 93% reported losses in their sample, the main reasons being categorized as follows: lack of time (10%), personal reasons (24%), attendance (41%), health (20%), not meeting criteria during the development of the intervention (17%), did not wish to continue (27%) and unspecified (34%). 58% of the studies recruited children from health centers, 27% from educational centers, and 51% from the community. The mean age was 9.3 years. Concerning gender distribution, 56% were women and 44% men. Regarding the prevalence of diseases, only two studies detailed this information: one included a child with asthma, and the other only healthy children. The main inclusion criterion (90%) was the BMI value corresponding to overweight or obesity, while in 3% of the studies it was the value of weight for height and 7% did not specify this information. 51% of the interventions were carried out in health centers, 13% in educational centers, 13% in university centers, and 13% in community centers. Of these studies, 13% did not specify this information ( Table 2).

Risk of Bias Assessment
The risk of bias analysis revealed that the distribution of biases classified as "low risk" or "unclear risk" was similar, except for detection bias, in which 66% of the studies presented "unclear risk", and reporting bias, in which 100% of the studies presented "low risk". Among the four types of bias, studies classified as "high risk" were only for "performance bias", where more than 10% (22%) of the studies presented "high risk" (Figure 2).

Main Characteristics of the Interventions
The studies presented as main result six variables related to childhood overweight and obesity: 38% reported BMI z-score, 34% reported BMI-SDS (standard deviation), 24% considered BMI (kg/m2) and 10% considered other variables, these being the percentage of body fat, body weight according to height and BMI-p (percentile). Regarding the main post-intervention result, 72% of the interventions presented significant changes, while 28% did not present significant post-intervention changes. Only nine studies analyzed the maintenance of this effect over time, obtaining statistically significant maintenance in 77% of cases. One of the inclusion criteria of 38% of the studies was that at least one of the parents was overweight or obese. In relation to the subjects who underwent the intervention, 4% included only parents, 6% only children, 52% included parents and children, and 34% included the entire family. In 59% of the studies, the duration of the intervention was equal to or less than 6 months, and in 41% it was equal to or greater than 12 months. 57% of the studies presented follow-up of post-intervention results. Of these, 43% followed up for a period equal to or less than 6 months, while 57% followed up for a period equal to or greater than 12 months. Regarding the professionals who participated in the interventions, 52% were nutritionists, 34% qualified professionals in exercise prescription (physiotherapists, physical education teachers, and personal trainers), 32% doctors, 27% psychologists, 27% other types of professional (interventionists, counselors or facilitators), 23% included a nurse, 2% a social worker and 9% did not specify this information. The interventions were based on four components: (1) physical activity, (2) education, (3) behavioral therapy, (4) and nutrition. The physical activity component was included in 86% of the interventions, being implemented in 55% of cases as recommendations (general exercise recommendations, promotion of an active life, and reduction of time spent in sedentary behaviors), in 16% of cases as physical exercise programs (structured training program including mainly aerobic exercise, circuit training, and spinning), and in 23% of cases as recreational activities (mainly games). Of the interventions. 55% included an education component, which was implemented mainly as healthy lifestyle recommendations focused on health promotion. Regarding behavioral therapy, 66% of the interventions included this component through cognitive-behavioral strategies: stimulus control, self-monitoring, coping strategies, dealing with problems, positive reinforcement, planning, goal setting, behavior skills, and parenting skills. Finally, 80% of the interventions included a nutrition component, which was implemented in 45% of cases as recommendations focused on healthy eating (such as increasing the intake of water, fruits, and vegetables and reducing the intake of high-calorie foods), in 34% of the cases as food plans using mainly the "traffic light system", and in 6% of cases healthy eating workshops were included (Table 3).

Effects of the Interventions on the Primary Outcome According to the Duration of the Intervention
Duration equal to or less than six months: Considering all types of interventions, 77% obtained significant effects for the primary outcome. All the interventions that included the physical activity component showed significant effects on the primary outcome with the recreational activities modality, while with the recommendations modality they obtained significant effects in 59% of cases, and 83% of cases using the physical exercise program modality. All the interventions that included education had significant effects on the primary outcome, whereas those that included behavioral therapy had significant effects in 65% of cases. Finally, 100% of the interventions that included a nutrition component showed significant effects on the primary outcome when using the diet program modality, 63% with the recommendation's modality, and 50% when using the healthy cooking workshop modality. Furthermore, considering that all the interventions included at least two components, they all presented significant effects when they included the physical activity component in the form of recreational activities. However, the interventions that combined physical activity (recommendations approach), behavioral therapy, and nutrition component (recommendations approach) only presented significant effects when they also included the education component, Duration equal or more than twelve months: of the interventions, 89% had significant effects on the primary outcome. 67% of interventions that included a physical activity component presented significant effects with the recreational activities modality, 100% with the recommendations modality, and 100% with the physical exercise program modality. Regarding interventions that included education, 92% presented significant effects on the result. The interventions that included behavioral therapy had significant effects in 91% of cases. Finally, 100% of the interventions that included a nutrition component showed significant effects on the primary outcome with the diet program modality, 75% with the recommendations modality, and 100% with the healthy cooking workshop modality. Furthermore, only two interventions (11%) did not present significant effects, having in common that both included the nutrition component in the modality of recommendations (Table 4).

Variables Studied in the Articles
The variables studied in the different studies were grouped into three categories: nutritional status, physical and health condition, and psychological variables.

Assessment Instruments
The instruments used in the studies were grouped into three categories: nutritional status, physical and health condition, and psychological variables.
Cardiovascular risk: 31% of the studies took blood samples, while 13% measured blood pressure with a manual blood pressure cuff, and 20% with an automatic sphygmomanometer (Figure 4e).

Psychological Variables
Ten instruments were used to assess these variables, and were sub-grouped into two categories: Psychological health: to assess this category, 6% of studies used the self-perceptions profile questionnaire for children (SPPC), 6% the children's depression scale (CDS), 6% the questionnaire of strengths and difficulties of the child reported by the parents (SDQ), 3% used the child and youth physical self-perception profile questionnaire (CY-PSPP), 3% the scale of intrinsic versus extrinsic orientation in the classroom, and 3% the multidimensional anxiety scale for children (MASC) (Figure 4f).
Health-related quality of life: of the studies, 6% used the pediatric quality of life questionnaire (PedsQL), 3% the child health questionnaire, Dutch version (CHQ-PF50) and 3% the quality of life questionnaire for children between 7 and 13 years old (KID-KINDL-R) (Figure 4g).

Discussion
The main finding of this review is that interventions that are effective for the management of childhood overweight and obesity are those that combine the components of physical activity, diet, education, and behavioral therapy, and that involve children and parents as participants in the intervention. Specifically, the most effective interventions with a duration equal to or less than six months included recreational activities, education, and nutritional programs. The most effective interventions lasting 12 months or more included physical activity recommendations, nutrition programs, physical exercise programs, and cooking classes.
For more details, the findings of this review are organized into four categories: the general characteristics of the studies, the assessment of the risk of bias, the main characteristics of the interventions, and the variables studied, as well as the assessment instruments used. Their importance and possible implications are discussed below.

General Characteristics of the Studies
In this review, 12 countries belonging to North America, Europe, and Asia were included, where interventions were developed for the management of obesity in children between 6 and 12 years of age with overweight, or obesity in geographic areas with extremely cold climates. Childhood overweight and obesity are a global problem [1,7], with a higher prevalence in regions where temperatures are lower [16,18]. Although 93% of the studies suffered losses in their sample, the average adherence was very high, 86%, which makes for highly relevant data, since the results of an intervention are dependent on the adherence of its participants. This is in contrast to what some authors state, that dropout at 6 months is close to 40% [56,57]. This high adherence could be related to the fact that the interventions had a large participation of parents, and that most of them included a behavioral therapy intervention, which is proposed as an effective tool to improve adherence to management programs of childhood obesity [58]. Regarding recruitment, this was carried out mainly from health centers. This could be related to the availability of the records in the health systems of developed countries, which in turn allow better access to potential participants. The main inclusion criterion was BMI (90%), surpassing other measures that could be more representative of overweight and obesity, such as the percentage of body fat. BMI, even though it is less specific, is the main tool for classifying overweight and obesity, perhaps because it is easy to access and low cost. BMI is the indicator recommended by the WHO to assess anthropometrically the population under 20 years of age, although it is essential to consider the degree of pubertal development before applying it [59,60]. 50% of the interventions took place in health centers, and only 13% in educational centers, although educational establishments are the space in which children spend a large part of their time during the day

Risk of Bias Assessment
The distribution of biases classified as "low risk" or "unclear risk" was similar, except for detection bias, which presented a 66% "unclear" risk. Detection bias is related to the blinding of the assessors to the study results, and although the Cochrane manual states that their blinding does not ensure success, lack of blinding could bias the study results [26]. In any case, the predominant classification of this type of bias, in particular, was "unclear risk" and not "high risk", which is associated with a lack of information regarding this bias on the part of the authors, rather than a possible bias in the results. Besides, although the studies had some "high risk" bias classification, the highest proportion was present in performance bias with 22%, and less than 10% in the other types of bias. This could be because all the studies have an experimental and randomized methodological design (RCT), which corresponds to the highest level of evidence in terms of quantitative studies.

Main Characteristics of the Interventions
Most of the studies considered BMI as the main outcome to assess the effects of the interventions. Although this is logical, since it was the main inclusion criterion to classify overweight and obesity, some authors question its use because it is less sensitive to weight change in children with higher body weight [61]. A change in BMI does not necessarily imply a decrease in body fat [62], but it is the indicator recommended by the WHO for anthropometric evaluation in children [59,60]. Most of the studies considered the participation of at least one parent during the intervention. Knowing the role that parents play in the development of obesity in their children, it is of great value that they can actively participate in interventions for its management [63,64]. The current gold standard treatment for the management of overweight and obesity in childhood is aimed at both children and parents [65,66], combining components such as physical activity, diet, education, and behavioral therapy [67]. Combined treatment, in addition to nutritional and physical activity intervention, should include behavioral management and education [68,69], to guarantee the maintenance of long-term effects through changes in behavior that promote sustained changes in lifestyle [58,70]. The effects of interventions that combine components have been documented both in the short term (≤6 months) and in the long term (≥12 months) [69,71]. The interventions that lasted 6 months or less presented significant effects in 77% of cases, and according to the components addressed the modalities that presented greater effectiveness were: in physical activity, recreational activities (100%), nutrition and diet program (100%), education (100%) and behavioral therapy (65%). Regarding the interventions that had a duration of at least 12 months, 89% presented significant changes in their primary effect, and according to the components addressed the modalities that presented greater effectiveness were: physical activity recommendations (100%), the modality of diet programs (100%), and the components of education (92%) and behavioral therapy (91%). It seems that the modality of physical activity recommendations, which consists of recommendations of intensity, time, and frequency, in addition to promoting active transport and reducing sedentary behaviors, has an important effect when the interventions are long-term, but not when the duration of the intervention is less than or equal to 6 months. Regarding the nutrition component, the healthy eating recommendations mode does not have as powerful an effect as the nutrition plans mode. Furthermore, the interventions that include the behavioral therapy component appear not to be as effective when combined with physical activity recommendations and healthy eating recommendations, as opposed to when combined with recreational activities and nutrition plans. It is important to emphasize that the treatment of obesity is not focused only on the reduction of body weight, and it is advisable to do a post-intervention follow-up to maintain the long-term effects [72]. Despite this, only 31% of the studies considered a post-intervention follow-up evaluation

Variables Studied and Assessment Instruments Used
The nutritional status variables were present in all studies, with BMI being the most studied variable despite its limitations, such as not allowing the distinction between fat mass and lean mass. It has also been observed that some children with normal BMI have fat mass values in the obesity range [73]. Only about 40% of the studies evaluated body composition through variables other than BMI. To do this, they used doubly indirect methods such as waist circumference and indirect methods such as bioimpedance and DXA, which some authors propose as a better alternative [74]. The physical health variables were considered in 76% of studies, with the level of physical activity being the most analyzed variable, recorded mainly by indirect methods such as questionnaires, with direct methods such as pedometers and accelerometers rarely used. This could be related to the fact that the physical activity component was approached mainly in the form of physical activity recommendations, rather than with physical exercise programs. Psychological variables were assessed in only 24% of the studies, with health-related quality of life being the most analyzed variable through questionnaires, surpassing other variables such as depression, anxiety, and self-esteem. Obesity is a disorder of multifactorial origin that affects multiple areas of human development, and we believe that the evaluation of psychological variables can provide very relevant data, since children with obesity have a very deteriorated image of themselves, feeling inferior and rejected, which is a drag on their social and psychological development in the short and long term [75,76].

Limitations and Contributions
A limitation of this study is not having found, in the literature, interventions carried out in other representative countries in areas of extreme cold climate such as Russia, or southern Chile and Argentina. However, studies from 12 countries on three different continents were included, allowing a broad view of the characteristics of interventions to combat childhood overweight and obesity in geographic areas of extreme cold climate. This study can serve as a guide to implement interventions in areas with these characteristics, knowing which ones are the most effective, concerning the components addressed and their duration.

Conclusions
The interventions that were effective in the management of childhood overweight and obesity were those that combined the components of physical activity, diet, education, and behavioral therapy and that involved both children and their parents as participants in the interventions. In interventions with a duration equal to or less than six months, the most effective included recreational activities, education and nutritional programs. In interventions lasting 12 months or more, the most effective included recommendations for physical activity, nutritional and physical exercise programs, and cooking classes. This demonstrates that the effectiveness of these programs requires a combination of components, with a multi-professional approach, aimed at implementing strategies that favor a healthy lifestyle in children, involving parents as active participants in the process. The results of this review serve as a reference to design future effective interventions in the management of childhood obesity in extremely cold climates.

Funding:
The authors received no financial support for the research, authorship, and/or publication of this article.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The datasets used and/or analyzed during the current study are avail-able from the corresponding author on reasonable request.

Conflicts of Interest:
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.