Motivations for weight loss in adolescents with overweight and obesity: a systematic review

The Meta-Analyzes of Observational Studies in Epidemiology (MOOSE) [11] and the Preferred Reporting Items for Systematic Reviews and Meta-Analyzes (PRISMA) [12] were used to plan, conduct, and compose the present systematic review.

The systematic review protocol was recorded in the International Prospective Register of Ongoing Systematic Reviews (PROSPERO) (<http://​www.​crd.​york.​ac.​uk/​PROSPERO/​>) under No. CRD42017056528.

The inclusion criteria were as follows: 1) Design: observational studies (cross-sectional and cohort); 2) Language: publications in Portuguese, English and/or Spanish; 3) Publication period: no restrictions; 4) Outcome: motivation for weight loss; 5) Population: adolescents aged 10 to 20 years. This age range was defined considering that it has been used in some studies on motivation and behaviors in adolescents [13, 14] and systematic reviews [15, 16]. Additionally, according to World Health Organization (WHO): “Age is often more appropriate for assessing and comparing biological changes (e.g. puberty), which are fairly universal” [17]. Thus, the use of a more expanded age range can be justified considering that behaviors, such as motivations for weight loss, can vary with the socio-cultural environment. Review articles were excluded.

Electronic searches were performed in the LILACS, PubMed, Scopus, and ADOLEC databases. In the LILACS and ADOLEC databases, the following combinations of terms were used: (motivation OR motivations OR “motivational factor” OR “motivational factors” OR motive OR motives OR reason OR reasons OR predictor) AND (“weight loss” OR “lose weight” OR “body-weight reduction”) AND (adolescent OR adolescents). For the PubMed and Scopus databases, the search strategy used was: (motivation OR motivations OR “motivating factor” OR “motivating factors” OR motive OR motives OR reason OR reasons OR predictor) AND (“lose weight” OR “losing weight” OR “weight loss” OR slimming) AND (adolescent OR adolescents).

To recover unpublished records, the thesis and dissertation banks of the following institutions were searched: The Coordination for the Improvement of Higher Education Personnel, the University of São Paulo, the Brazilian Institute of Information in Science and Technology. The OpenGrey database (<http://​www.​opengrey.​eu/​>), which specializes in grey literature, was also searched, and a manual search was performed.

The first author (D.S.) triaged the retrieved records, while another author (S.L.) reviewed the selected studies. After identifying the studies that would be included in the review, the researchers collected the following data: journal, language, year of publication, authorship, country of origin, number and age of participants, method of nutritional status classification, instrument for assessing motivations and main motivation for weight loss. The motivations for weight loss were compared across studies, with consideration of the similarity among the motivations. That is, studies that described similar motivations related to appearance were considered as reporting the same motivation.

Two researchers assessed the methodological quality of the studies included in the review using the Agency for Healthcare Research and Quality (AHRQ) Methodology Checklist for Cross-Sectional/Prevalence Studies (http://​www.​ncbi.​nlm.​nih.​gov/​books/​NBK35156/​) [18]. The AHRQ checklist consists of 11 items that can be answered with either “yes”, “no”, or “unclear”. We classified studies with 8–11 “yes” responses as “high quality”; those with 4–7 were “moderate quality”, and those with 0–3 were “low quality”.

The search in the PubMed, Scopus, LILACS, and ADOLEC databases retrieved 1635 records. In the manual search, 13 articles were retrieved for a total of 1648 records. After reading the titles and the abstracts, we excluded 1616 abstracts. Of the 32 abstracts selected, five were duplicates. The remaining 27 were assessed for eligibility. After the inclusion and exclusion criteria were assessed, six articles were included in the present review; all were cross-sectional studies [19‐24]. Figure 1 presents the flowchart for the selection of the studies.

The methodological quality and characteristics of the studies are presented in Table 1. Five studies were classified as having moderate quality because they received a “yes” response for 4–7 items of the AHRQ checklist [19‐21, 23, 24]. One study received a “yes” response for eight items and was classified as high quality [22].

Two studies were conducted in the United States of America [19, 20], one in Thailand [21], one in Sweden [22], one in the United Kingdom [23], and one in Brazil [24]. The range of sample sizes varied between 12 [23] and 167 [21] adolescents; whereas the age range went from 11 [23] to 20 [19] years.

The most commonly used diagnostic criterion for overweight and obesity among the studies was that of the Center for Disease Control (percentile of BMI by age), which was used in two studies [19, 24]. Other studies established different criteria and, in particular, did not define (extreme) obesity uniformly. Morinder et al. [22], for example, used the International Criteria of the Obesity Task Force, whereas In-Iw et al. [21] applied the following criteria: slightly obese - BMI ≥ 25 kg/m² combined with a weight between 120 and 140%, and moderately-severely obese - BMI ≥ 25 kg/m² combined with weight for body size percentage > 140%. Reece et al. [23] defined overweight as a BMI percentile > 91, and the study by Taylor et al. [20] classified severely obese as a weight for height percentage ≥ 150% [15]. One study included only adolescents with overweight [19], four included only adolescents with obesity [20‐22, 24], and one included both [23].

The instruments for assessing and the motivations for weight loss among adolescents with overweight and obesity are presented in Table 2 and Fig. 2. An interview with open questions was the most frequent instrument, used in four studies [20, 22‐24]. Two studies used questionnaires with open questions [19, 21]. An example of an open question used in the studies would be “What are your personal motivations for losing weight?”, which was used in the study by Lofrano-Prado et al. [24].

In the six studies included in the present review, 17 different motivations for weight loss in adolescents with overweight and obesity were identified [19‐24]. The desire for health improvements was the main one, observed in four studies [19‐22], of which three included exclusively adolescents with obesity [20‐22]. In three studies, the participants stated appearance reasons [19, 21, 24]. Improvement in self-esteem was given as a motivation in three studies [19, 22, 24], while avoidance of bullying was also mentioned in three other ones studies [20, 23, 24]. The motivations to be healthier [22, 24], to be in better fitness/shape [22, 24], to have peer acceptance and/or greater social integration [19, 23], and to fit into “normal size”, more elegant clothing [20, 24] were identified in two studies each. The following motivations were mentioned in one study each: being more attractive to the opposite sex [21], doing things for enjoyment [19], facing major changes in life (school or city) [19], personal motivation [19], celebrating the 15th birthday [20], increased mobility, practicing physical activities [20], feeling good [22], accepting oneself [22], and improving quality of life [24].